idnits 2.17.1 draft-ietf-behave-rfc3489bis-16.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 20. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 2315. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 2326. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 2333. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 2339. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust Copyright Line does not match the current year == The document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (July 2, 2008) is 5771 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) ** Obsolete normative reference: RFC 2818 (Obsoleted by RFC 9110) ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Obsolete normative reference: RFC 2617 (Obsoleted by RFC 7235, RFC 7615, RFC 7616, RFC 7617) ** Obsolete normative reference: RFC 2988 (Obsoleted by RFC 6298) ** Downref: Normative reference to an Informational RFC: RFC 2104 -- Possible downref: Non-RFC (?) normative reference: ref. 'ITU.V42.2002' ** Downref: Normative reference to an Informational RFC: RFC 1321 ** Obsolete normative reference: RFC 4013 (Obsoleted by RFC 7613) -- Obsolete informational reference (is this intentional?): RFC 2616 (Obsoleted by RFC 7230, RFC 7231, RFC 7232, RFC 7233, RFC 7234, RFC 7235) -- Obsolete informational reference (is this intentional?): RFC 3489 (Obsoleted by RFC 5389) == Outdated reference: A later version (-16) exists of draft-ietf-behave-turn-06 == Outdated reference: A later version (-20) exists of draft-ietf-sip-outbound-11 == Outdated reference: A later version (-08) exists of draft-ietf-behave-nat-behavior-discovery-02 == Outdated reference: A later version (-16) exists of draft-ietf-mmusic-ice-tcp-05 -- Obsolete informational reference (is this intentional?): RFC 5226 (Obsoleted by RFC 8126) Summary: 8 errors (**), 0 flaws (~~), 6 warnings (==), 12 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 BEHAVE Working Group J. Rosenberg 3 Internet-Draft Cisco 4 Obsoletes: 3489 (if approved) R. Mahy 5 Intended status: Standards Track Plantronics 6 Expires: January 3, 2009 P. Matthews 7 Avaya 8 D. Wing 9 Cisco 10 July 2, 2008 12 Session Traversal Utilities for (NAT) (STUN) 13 draft-ietf-behave-rfc3489bis-16 15 Status of this Memo 17 By submitting this Internet-Draft, each author represents that any 18 applicable patent or other IPR claims of which he or she is aware 19 have been or will be disclosed, and any of which he or she becomes 20 aware will be disclosed, in accordance with Section 6 of BCP 79. 22 Internet-Drafts are working documents of the Internet Engineering 23 Task Force (IETF), its areas, and its working groups. Note that 24 other groups may also distribute working documents as Internet- 25 Drafts. 27 Internet-Drafts are draft documents valid for a maximum of six months 28 and may be updated, replaced, or obsoleted by other documents at any 29 time. It is inappropriate to use Internet-Drafts as reference 30 material or to cite them other than as "work in progress." 32 The list of current Internet-Drafts can be accessed at 33 http://www.ietf.org/ietf/1id-abstracts.txt. 35 The list of Internet-Draft Shadow Directories can be accessed at 36 http://www.ietf.org/shadow.html. 38 This Internet-Draft will expire on January 3, 2009. 40 Copyright Notice 42 Copyright (C) The IETF Trust (2008). 44 Abstract 46 Session Traversal Utilities for NAT (STUN) is a protocol that serves 47 as a tool for other protocols in dealing with NAT traversal. It can 48 be used by an endpoint to determine the IP address and port allocated 49 to it by a NAT. It can also be used to check connectivity between 50 two endpoints, and as a keep-alive protocol to maintain NAT bindings. 51 STUN works with many existing NATs, and does not require any special 52 behavior from them. 54 STUN is not a NAT traversal solution by itself. Rather, it is a tool 55 to be used in the context of a NAT traversal solution. This is an 56 important change from the previous version of this specification (RFC 57 3489), which presented STUN as a complete solution. 59 This document obsoletes RFC 3489. 61 Table of Contents 63 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 64 2. Evolution from RFC 3489 . . . . . . . . . . . . . . . . . . . 4 65 3. Overview of Operation . . . . . . . . . . . . . . . . . . . . 5 66 4. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 8 67 5. Definitions . . . . . . . . . . . . . . . . . . . . . . . . . 8 68 6. STUN Message Structure . . . . . . . . . . . . . . . . . . . . 10 69 7. Base Protocol Procedures . . . . . . . . . . . . . . . . . . . 12 70 7.1. Forming a Request or an Indication . . . . . . . . . . . 12 71 7.2. Sending the Request or Indication . . . . . . . . . . . . 13 72 7.2.1. Sending over UDP . . . . . . . . . . . . . . . . . . . 13 73 7.2.2. Sending over TCP or TLS-over-TCP . . . . . . . . . . . 14 74 7.3. Receiving a STUN Message . . . . . . . . . . . . . . . . 16 75 7.3.1. Processing a Request . . . . . . . . . . . . . . . . . 17 76 7.3.1.1. Forming a Success or Error Response . . . . . . . 18 77 7.3.1.2. Sending the Success or Error Response . . . . . . 18 78 7.3.2. Processing an Indication . . . . . . . . . . . . . . . 19 79 7.3.3. Processing a Success Response . . . . . . . . . . . . 19 80 7.3.4. Processing an Error Response . . . . . . . . . . . . . 19 81 8. FINGERPRINT Mechanism . . . . . . . . . . . . . . . . . . . . 20 82 9. DNS Discovery of a Server . . . . . . . . . . . . . . . . . . 21 83 10. Authentication and Message-Integrity Mechanisms . . . . . . . 22 84 10.1. Short-Term Credential Mechanism . . . . . . . . . . . . . 22 85 10.1.1. Forming a Request or Indication . . . . . . . . . . . 22 86 10.1.2. Receiving a Request or Indication . . . . . . . . . . 23 87 10.1.3. Receiving a Response . . . . . . . . . . . . . . . . . 24 88 10.2. Long-term Credential Mechanism . . . . . . . . . . . . . 24 89 10.2.1. Forming a Request . . . . . . . . . . . . . . . . . . 25 90 10.2.1.1. First Request . . . . . . . . . . . . . . . . . . 25 91 10.2.1.2. Subsequent Requests . . . . . . . . . . . . . . . 25 92 10.2.2. Receiving a Request . . . . . . . . . . . . . . . . . 25 93 10.2.3. Receiving a Response . . . . . . . . . . . . . . . . . 26 94 11. ALTERNATE-SERVER Mechanism . . . . . . . . . . . . . . . . . . 27 95 12. Backwards Compatibility with RFC 3489 . . . . . . . . . . . . 28 96 12.1. Changes to Client Processing . . . . . . . . . . . . . . 28 97 12.2. Changes to Server Processing . . . . . . . . . . . . . . 29 98 13. Basic Server Behavior . . . . . . . . . . . . . . . . . . . . 29 99 14. STUN Usages . . . . . . . . . . . . . . . . . . . . . . . . . 30 100 15. STUN Attributes . . . . . . . . . . . . . . . . . . . . . . . 31 101 15.1. MAPPED-ADDRESS . . . . . . . . . . . . . . . . . . . . . 32 102 15.2. XOR-MAPPED-ADDRESS . . . . . . . . . . . . . . . . . . . 32 103 15.3. USERNAME . . . . . . . . . . . . . . . . . . . . . . . . 33 104 15.4. MESSAGE-INTEGRITY . . . . . . . . . . . . . . . . . . . . 34 105 15.5. FINGERPRINT . . . . . . . . . . . . . . . . . . . . . . . 35 106 15.6. ERROR-CODE . . . . . . . . . . . . . . . . . . . . . . . 35 107 15.7. REALM . . . . . . . . . . . . . . . . . . . . . . . . . . 37 108 15.8. NONCE . . . . . . . . . . . . . . . . . . . . . . . . . . 37 109 15.9. UNKNOWN-ATTRIBUTES . . . . . . . . . . . . . . . . . . . 37 110 15.10. SERVER . . . . . . . . . . . . . . . . . . . . . . . . . 38 111 15.11. CLIENT . . . . . . . . . . . . . . . . . . . . . . . . . 38 112 15.12. ALTERNATE-SERVER . . . . . . . . . . . . . . . . . . . . 38 113 16. Security Considerations . . . . . . . . . . . . . . . . . . . 39 114 16.1. Attacks against the Protocol . . . . . . . . . . . . . . 39 115 16.1.1. Outside Attacks . . . . . . . . . . . . . . . . . . . 39 116 16.1.2. Inside Attacks . . . . . . . . . . . . . . . . . . . . 40 117 16.2. Attacks Affecting the Usage . . . . . . . . . . . . . . . 40 118 16.2.1. Attack I: DDoS Against a Target . . . . . . . . . . . 41 119 16.2.2. Attack II: Silencing a Client . . . . . . . . . . . . 41 120 16.2.3. Attack III: Assuming the Identity of a Client . . . . 41 121 16.2.4. Attack IV: Eavesdropping . . . . . . . . . . . . . . . 41 122 16.3. Hash Agility Plan . . . . . . . . . . . . . . . . . . . . 42 123 17. IAB Considerations . . . . . . . . . . . . . . . . . . . . . . 42 124 18. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 42 125 18.1. STUN Methods Registry . . . . . . . . . . . . . . . . . . 42 126 18.2. STUN Attribute Registry . . . . . . . . . . . . . . . . . 43 127 18.3. STUN Error Code Registry . . . . . . . . . . . . . . . . 44 128 18.4. STUN UDP and TCP Port Numbers . . . . . . . . . . . . . . 44 129 19. Changes Since RFC 3489 . . . . . . . . . . . . . . . . . . . . 44 130 20. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 46 131 21. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 46 132 22. References . . . . . . . . . . . . . . . . . . . . . . . . . . 46 133 22.1. Normative References . . . . . . . . . . . . . . . . . . 46 134 22.2. Informational References . . . . . . . . . . . . . . . . 47 135 Appendix A. C Snippet to Determine STUN Message Types . . . . . . 49 136 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 49 137 Intellectual Property and Copyright Statements . . . . . . . . . . 51 139 1. Introduction 141 The protocol defined in this specification, Session Traversal 142 Utilities for NAT, provides a tool for dealing with NATs. It 143 provides a means for an endpoint to determine the IP address and port 144 allocated by a NAT that corresponds to its private IP address and 145 port. It also provides a way for an endpoint to keep a NAT binding 146 alive. With some extensions, the protocol can be used to do 147 connectivity checks between two endpoints [I-D.ietf-mmusic-ice], or 148 to relay packets between two endpoints [I-D.ietf-behave-turn]. 150 In keeping with its tool nature, this specification defines an 151 extensible packet format, defines operation over several transport 152 protocols, and provides for two forms of authentication. 154 STUN is intended to be used in context of one or more NAT traversal 155 solutions. These solutions are known as STUN usages. Each usage 156 describes how STUN is utilized to achieve the NAT traversal solution. 157 Typically, a usage indicates when STUN messages get sent, which 158 optional attributes to include, what server is used, and what 159 authentication mechanism is to be used. Interactive Connectivity 160 Establishment (ICE) [I-D.ietf-mmusic-ice] is one usage of STUN. SIP 161 Outbound [I-D.ietf-sip-outbound] is another usage of STUN. In some 162 cases, a usage will require extensions to STUN. A STUN extension can 163 be in the form of new methods, attributes, or error response codes. 164 More information on STUN usages can be found in Section 14. 166 2. Evolution from RFC 3489 168 STUN was originally defined in RFC 3489 [RFC3489]. That 169 specification, sometimes referred to as "classic STUN", represented 170 itself as a complete solution to the NAT traversal problem. In that 171 solution, a client would discover whether it was behind a NAT, 172 determine its NAT type, discover its IP address and port on the 173 public side of the outermost NAT, and then utilize that IP address 174 and port within the body of protocols, such as the Session Initiation 175 Protocol (SIP) [RFC3261]. However, experience since the publication 176 of RFC 3489 has found that classic STUN simply does not work 177 sufficiently well to be a deployable solution. The address and port 178 learned through classic STUN are sometimes usable for communications 179 with a peer, and sometimes not. Classic STUN provided no way to 180 discover whether it would, in fact, work or not, and it provided no 181 remedy in cases where it did not. Furthermore, classic STUN's 182 algorithm for classification of NAT types was found to be faulty, as 183 many NATs did not fit cleanly into the types defined there. 185 Classic STUN also had a security vulnerability - attackers could 186 provide the client with incorrect mapped addresses under certain 187 topologies and constraints, and this was fundamentally not solvable 188 through any cryptographic means. Though this problem remains with 189 this specification, those attacks are now mitigated through the use 190 of more complete solutions that make use of STUN. 192 For these reasons, this specification obsoletes RFC 3489, and instead 193 describes STUN as a tool that is utilized as part of a complete NAT 194 traversal solution. ICE [I-D.ietf-mmusic-ice] is a complete NAT 195 traversal solution for protocols based on the offer/answer [RFC3264] 196 methodology, such as SIP. SIP Outbound [I-D.ietf-sip-outbound] is a 197 complete solution for traversal of SIP signaling, and it uses STUN in 198 a very different way. Though it is possible that a protocol may be 199 able to use STUN by itself (classic STUN) as a traversal solution, 200 such usage is not described here and is strongly discouraged for the 201 reasons described above. 203 The on-the-wire protocol described here is changed only slightly from 204 classic STUN. The protocol now runs over TCP in addition to UDP. 205 Extensibility was added to the protocol in a more structured way. A 206 magic-cookie mechanism for demultiplexing STUN with application 207 protocols was added by stealing 32 bits from the 128 bit transaction 208 ID defined in RFC 3489, allowing the change to be backwards 209 compatible. Mapped addresses are encoded using a new exclusive-or 210 format. There are other, more minor changes. See Section 19 for a 211 more complete listing. 213 Due to the change in scope, STUN has also been renamed from "Simple 214 Traversal of UDP Through NAT" to "Session Traversal Utilities for 215 NAT". The acronym remains STUN, which is all anyone ever remembers 216 anyway. 218 3. Overview of Operation 220 This section is descriptive only. 222 /-----\ 223 // STUN \\ 224 | Server | 225 \\ // 226 \-----/ 228 +--------------+ Public Internet 229 ................| NAT 2 |....................... 230 +--------------+ 232 +--------------+ Private NET 2 233 ................| NAT 1 |....................... 234 +--------------+ 236 /-----\ 237 // STUN \\ 238 | Client | 239 \\ // Private NET 1 240 \-----/ 242 Figure 1: One possible STUN Configuration 244 One possible STUN configuration is shown in Figure 1. In this 245 configuration, there are two entities (called STUN agents) that 246 implement the STUN protocol. The lower agent in the figure is the 247 client, and is connected to private network 1. This network connects 248 to private network 2 through NAT 1. Private network 2 connects to 249 the public Internet through NAT 2. The upper agent in the figure is 250 the server, and resides on the public Internet. 252 STUN is a client-server protocol. It supports two types of 253 transactions. One is a request/response transaction in which a 254 client sends a request to a server, and the server returns a 255 response. The second is an indication transaction in which either 256 agent - client or server - sends an indication which generates no 257 response. Both types of transactions include a transaction ID, which 258 is a randomly selected 96-bit number. For request/response 259 transactions, this transaction ID allows the client to associate the 260 response with the request that generated it; for indications, this 261 simply serves as a debugging aid. 263 All STUN messages start with a fixed header that includes a method, a 264 class, and the transaction ID. The method indicates which of the 265 various requests or indications this is; this specification defines 266 just one method, Binding, but other methods are expected to be 267 defined in other documents. The class indicates whether this is a 268 request, a success response, an error response, or an indication. 269 Following the fixed header comes zero or more attributes, which are 270 type-length-value extensions that convey additional information for 271 the specific message. 273 This document defines a single method called Binding. The Binding 274 method can be used either in request/response transactions or in 275 indication transactions. When used in request/response transactions, 276 the Binding method can be used to determine the particular "binding" 277 a NAT has allocated to a STUN client. When used in either request/ 278 response or in indication transactions, the Binding method can also 279 be used to keep these "bindings" alive. 281 In the Binding request/response transaction, a Binding Request is 282 sent from a STUN client to a STUN server. When the Binding Request 283 arrives at the STUN server, it may have passed through one or more 284 NATs between the STUN client and the STUN server (in Figure 1, there 285 were two such NATs). As the Binding Request message passes through a 286 NAT, the NAT will modify the source transport address (that is, the 287 source IP address and the source port) of the packet. As a result, 288 the source transport address of the request received by the server 289 will be the public IP address and port created by the NAT closest to 290 the server. This is called a reflexive transport address. The STUN 291 server copies that source transport address into an XOR-MAPPED- 292 ADDRESS attribute in the STUN Binding Response and sends the Binding 293 Response back to the STUN client. As this packet passes back through 294 a NAT, the NAT will modify the destination transport address in the 295 IP header, but the transport address in the XOR-MAPPED-ADDRESS 296 attribute within the body of the STUN response will remain untouched. 297 In this way, the client can learn its reflexive transport address 298 allocated by the outermost NAT with respect to the STUN server. 300 In some usages, STUN must be multiplexed with other protocols (e.g., 301 [I-D.ietf-mmusic-ice], [I-D.ietf-sip-outbound]). In these usages, 302 there must be a way to inspect a packet and determine if it is a STUN 303 packet or not. STUN provides three fields in the STUN header with 304 fixed values that can be used for this purpose. If this is not 305 sufficient, then STUN packets can also contain a FINGERPRINT value 306 which can further be used to distinguish the packets. 308 STUN defines a set of optional procedures that a usage can decide to 309 use, called mechanisms. These mechanisms include DNS discovery, a 310 redirection technique to an alternate server, a fingerprint attribute 311 for demultiplexing, and two authentication and message integrity 312 exchanges. The authentication mechanisms revolve around the use of a 313 username, password, and message-integrity value. Two authentication 314 mechanisms, the long-term credential mechanism and the short-term 315 credential mechanism, are defined in this specification. Each usage 316 specifies the mechanisms allowed with that usage. 318 In the long-term credential mechanism, the client and server share a 319 pre-provisioned username and password and perform a digest challenge/ 320 response exchange inspired by (but differing in details) to the one 321 defined for HTTP [RFC2617]. In the short-term credential mechanism, 322 the client and the server exchange a username and password through 323 some out-of-band method prior to the STUN exchange. For example, in 324 the ICE usage [I-D.ietf-mmusic-ice] the two endpoints use out-of-band 325 signaling to exchange a username and password. These are used to 326 integrity protect and authenticate the request and response. There 327 is no challenge or nonce used. 329 4. Terminology 331 In this document, the key words "MUST", "MUST NOT", "REQUIRED", 332 "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", 333 and "OPTIONAL" are to be interpreted as described in BCP 14, RFC 2119 334 [RFC2119] and indicate requirement levels for compliant STUN 335 implementations. 337 5. Definitions 339 STUN Agent: An entity that implements the STUN protocol. The entity 340 can either be a STUN client or a STUN server. 342 STUN Client: A STUN client is an entity that sends STUN requests, 343 and receives STUN responses. STUN clients can also send 344 indications. In this specification, the terms STUN client and 345 client are synonymous. 347 STUN Server: A STUN server is an entity that receives STUN requests 348 and sends STUN responses. A STUN server can also send 349 indications. In this specification, the terms STUN server and 350 server are synonymous. 352 Transport Address: The combination of an IP address and port number 353 (such as a UDP or TCP port number). 355 Reflexive Transport Address: A transport address learned by a client 356 that identifies that client as seen by another host on an IP 357 network, typically a STUN server. When there is an intervening 358 NAT between the client and the other host, the reflexive transport 359 address represents the mapped address allocated to the client on 360 the public side of the NAT. Reflexive transport addresses are 361 learned from the mapped address attribute (MAPPED-ADDRESS or XOR- 362 MAPPED-ADDRESS) in STUN responses. 364 Mapped Address: Same meaning as Reflexive Address. This term is 365 retained only for for historic reasons and due to the naming of 366 the MAPPED-ADDRESS and XOR-MAPPED-ADDRESS attributes. 368 Long Term Credential: A username and associated password that 369 represent a shared secret between client and server. Long term 370 credentials are generally granted to the client when a subscriber 371 enrolls in a service and persist until the subscriber leaves the 372 service or explicitly changes the credential. 374 Long Term Password: The password from a long term credential. 376 Short Term Credential: A temporary username and associated password 377 which represent a shared secret between client and server. Short 378 term credentials are obtained through some kind of protocol 379 mechanism between the client and server, preceding the STUN 380 exchange. A short term credential has an explicit temporal scope, 381 which may be based on a specific amount of time (such as 5 382 minutes) or on an event (such as termination of a SIP dialog). 383 The specific scope of a short term credential is defined by the 384 application usage. 386 Short Term Password: The password component of a short term 387 credential. 389 STUN Indication: A STUN message that does not receive a response 391 Attribute: The STUN term for a Type-Length-Value (TLV) object that 392 can be added to a STUN message. Attributes are divided into two 393 types: comprehension-required and comprehension-optional. STUN 394 agents can safely ignore comprehension-optional attributes they 395 don't understand, but cannot successfully process a message if it 396 contains comprehension-required attributes that are not 397 understood. 399 RTO: Retransmission TimeOut, which defines the initial period of 400 time between transmission of a request and the first retransmit of 401 that request. 403 6. STUN Message Structure 405 STUN messages are encoded in binary using network-oriented format 406 (most significant byte or octet first, also commonly known as big- 407 endian). The transmission order is described in detail in Appendix B 408 of RFC791 [RFC0791]. Unless otherwise noted, numeric constants are 409 in decimal (base 10). 411 All STUN messages MUST start with a 20-byte header followed by zero 412 or more Attributes. The STUN header contains a STUN message type, 413 magic cookie, transaction ID, and message length. 415 0 1 2 3 416 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 417 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 418 |0 0| STUN Message Type | Message Length | 419 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 420 | Magic Cookie | 421 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 422 | | 423 | Transaction ID (96 bits) | 424 | | 425 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 427 Figure 2: Format of STUN Message Header 429 The most significant two bits of every STUN message MUST be zeroes. 430 This can be used to differentiate STUN packets from other protocols 431 when STUN is multiplexed with other protocols on the same port. 433 The message type defines the message class (request, success 434 response, failure response, or indication) and the message method 435 (the primary function) of the STUN message. Although there are four 436 message classes, there are only two types of transactions in STUN: 437 request/response transactions (which consist of a request message and 438 a response message), and indication transactions (which consists of a 439 single indication message). Response classes are split into error 440 and success responses to aid in quickly processing the STUN message. 442 The message type field is decomposed further into the following 443 structure: 445 0 1 446 2 3 4 5 6 7 8 9 0 1 2 3 4 5 448 +--+--+-+-+-+-+-+-+-+-+-+-+-+-+ 449 |M |M |M|M|M|C|M|M|M|C|M|M|M|M| 450 |11|10|9|8|7|1|6|5|4|0|3|2|1|0| 451 +--+--+-+-+-+-+-+-+-+-+-+-+-+-+ 453 Figure 3: Format of STUN Message Type Field 455 Here the bits in the message type field are shown as most-significant 456 (M11) through least-significant (M0). M11 through M0 represent a 12- 457 bit encoding of the method. C1 and C0 represent a 2 bit encoding of 458 the class. A class of 0b00 is a Request, a class of 0b01 is an 459 indication, a class of 0b10 is a success response, and a class of 460 0b11 is an error response. This specification defines a single 461 method, Binding. The method and class are orthogonal, so that for 462 each method, a request, success response, error response and 463 indication are defined for that method. 465 For example, a Binding Request has class=0b00 (request) and 466 method=0b000000000001 (Binding), and is encoded into the first 16 467 bits as 0x0001. A Binding response has class=0b10 (success response) 468 and method=0b000000000001, and is encoded into the first 16 bits as 469 0x0101. 471 Note: This unfortunate encoding is due to assignment of values in 472 [RFC3489] which did not consider encoding Indications, Success, 473 and Errors using bit fields. 475 The magic cookie field MUST contain the fixed value 0x2112A442 in 476 network byte order. In RFC 3489 [RFC3489], this field was part of 477 the transaction ID; placing the magic cookie in this location allows 478 a server to detect if the client will understand certain attributes 479 that were added in this revised specification. In addition, it aids 480 in distinguishing STUN packets from packets of other protocols when 481 STUN is multiplexed with those other protocols on the same port. 483 The transaction ID is a 96 bit identifier, used to uniquely identify 484 STUN transactions. For request/response transactions, the 485 transaction ID is chosen by the STUN client for the request and 486 echoed by the server in the response. For indications, it is chosen 487 by the agent sending the indication. It primarily serves to 488 correlate requests with responses, though it also plays a small role 489 in helping to prevent certain types of attacks. As such, the 490 transaction ID MUST be uniformly and randomly chosen from the 491 interval 0 .. 2**96-1. Resends of the same request reuse the same 492 transaction ID, but the client MUST choose a new transaction ID for 493 new transactions unless the new request is bit-wise identical to the 494 previous request and sent from the same transport address to the same 495 IP address. Success and error responses MUST carry the same 496 transaction ID as their corresponding request. When an agent is 497 acting as a STUN server and STUN client on the same port, the 498 transaction IDs in requests sent by the agent have no relationship to 499 the transaction IDs in requests received by the agent. 501 The message length MUST contain the size, in bytes, of the message 502 not including the 20 byte STUN header. Since all STUN attributes are 503 padded to a multiple of four bytes, the last two bits of this field 504 are always zero. This provides another way to distinguish STUN 505 packets from packets of other protocols. 507 Following the STUN fixed portion of the header are zero or more 508 attributes. Each attribute is TLV (type-length-value) encoded. The 509 details of the encoding, and of the attributes themselves is given in 510 Section 15. 512 7. Base Protocol Procedures 514 This section defines the base procedures of the STUN protocol. It 515 describes how messages are formed, how they are sent, and how they 516 are processed when they are received. It also defines the detailed 517 processing of the Binding method. Other sections in this document 518 describe optional procedures that a usage may elect to use in certain 519 situations. Other documents may define other extensions to STUN, by 520 adding new methods, new attributes, or new error response codes. 522 7.1. Forming a Request or an Indication 524 When formulating a request or indication message, the agent MUST 525 follow the rules in Section 6 when creating the header. In addition, 526 the message class MUST be either "Request" or "Indication" (as 527 appropriate), and the method must be either Binding or some method 528 defined in another document. 530 The agent then adds any attributes specified by the method or the 531 usage. For example, some usages may specify that the agent use an 532 authentication method (Section 10) or the FINGERPRINT attribute 533 (Section 8). In addition, the client SHOULD add a CLIENT attribute 534 to the message. 536 For the Binding method with no authentication, no attributes are 537 required unless the usage specifies otherwise. 539 All STUN messages sent over UDP SHOULD be less than the path MTU, if 540 known. If the path MTU is unknown, messages SHOULD be the smaller of 541 576 bytes and the first-hop MTU for IPv4 [RFC1122] and 1280 bytes for 542 IPv6 [RFC2460]. This value corresponds to the overall size of the IP 543 packet. Consequently, for IPv4, the actual STUN message would need 544 to be less than 548 bytes (576 minus 20 bytes IP header, minus 8 byte 545 UDP header, assuming no IP options are used). STUN provides no 546 ability to handle the case where the request is under the MTU but the 547 response would be larger than the MTU. It is not envisioned that 548 this limitation will be an issue for STUN. The MTU limitation is a 549 SHOULD, and not a MUST, to account for cases where STUN itself is 550 being used to probe for MTU characteristics 551 [I-D.ietf-behave-nat-behavior-discovery]. Outside of this or similar 552 applications, the MTU constraint MUST be followed. 554 7.2. Sending the Request or Indication 556 The agent then sends the request or indication. This document 557 specifies how to send STUN messages over UDP, TCP, or TLS-over-TCP; 558 other transport protocols may be added in the future. The STUN usage 559 must specify which transport protocol is used, and how the agent 560 determines the IP address and port of the recipient. Section 9 561 describes a DNS-based method of determining the IP address and port 562 of a server which a usage may elect to use. STUN may be used with 563 anycast addresses, but only with UDP and in usages where 564 authentication is not used. 566 At any time, a client MAY have multiple outstanding STUN requests 567 with the same STUN server (that is, multiple transactions in 568 progress, with different transaction ids). Absent other limits to 569 the rate of new transactions (such as those specified by ICE for 570 connectivity checks), a client SHOULD space new transactions to a 571 server by RTO and SHOULD limit itself to ten outstanding transactions 572 to the same server. 574 7.2.1. Sending over UDP 576 When running STUN over UDP it is possible that the STUN message might 577 be dropped by the network. Reliability of STUN request/response 578 transactions is accomplished through retransmissions of the request 579 message by the client application itself. STUN indications are not 580 retransmitted; thus indication transactions over UDP are not 581 reliable. 583 A client SHOULD retransmit a STUN request message starting with an 584 interval of RTO ("Retransmission TimeOut"), doubling after each 585 retransmission. The RTO is an estimate of the round-trip-time, and 586 is computed as described in RFC 2988 [RFC2988], with two exceptions. 587 First, the initial value for RTO SHOULD be configurable (rather than 588 the 3s recommended in RFC 2988) and SHOULD be greater than 500ms. 589 The exception cases for this SHOULD are when other mechanisms are 590 used to derive congestion thresholds (such as the ones defined in ICE 591 for fixed rate streams), or when STUN is used in non-Internet 592 environments with known network capacities. In fixed-line access 593 links, a value of 500ms is RECOMMENDED. Secondly, the value of RTO 594 MUST NOT be rounded up to the nearest second. Rather, a 1ms accuracy 595 MUST be maintained. As with TCP, the usage of Karn's algorithm is 596 RECOMMENDED [KARN87]. When applied to STUN, it means that RTT 597 estimates SHOULD NOT be computed from STUN transactions which result 598 in the retransmission of a request. 600 The value for RTO SHOULD be cached by a client after the completion 601 of the transaction, and used as the starting value for RTO for the 602 next transaction to the same server (based on equality of IP 603 address). The value SHOULD be considered stale and discarded after 604 10 minutes. 606 Retransmissions continue until a response is received, or until a 607 total of Rc requests have been sent. Rc SHOULD be configurable and 608 SHOULD have a default of 7. If, after the last request, a duration 609 equal to Rm times the RTO has passed without a response (providing 610 ample time to get a response if only this final request actually 611 succeeds), the client SHOULD consider the transaction to have failed. 612 Rm SHOULD be configurable and SHOULD have a default of 16. A STUN 613 transaction over UDP is also considered failed if there has been a 614 hard ICMP error [RFC1122]. For example, assuming an RTO of 500ms, 615 requests would be sent at times 0ms, 500ms, 1500ms, 3500ms, 7500ms, 616 15500ms, and 31500ms. If the client has not received a response 617 after 39500ms, the client will consider the transaction to have timed 618 out. 620 7.2.2. Sending over TCP or TLS-over-TCP 622 For TCP and TLS-over-TCP, the client opens a TCP connection to the 623 server. 625 In some usages of STUN, STUN is sent as the only protocol over the 626 TCP connection. In this case, it can be sent without the aid of any 627 additional framing or demultiplexing. In other usages, or with other 628 extensions, it may be multiplexed with other data over a TCP 629 connection. In that case, STUN MUST be run on top of some kind of 630 framing protocol, specified by the usage or extension, which allows 631 for the agent to extract complete STUN messages and complete 632 application layer messages. The STUN service running on the well 633 known port or ports discovered through the the DNS procedures in 634 Section 9 is for STUN alone, and not for STUN multiplexed with other 635 data. Consequently, no framing protocols are used in connections to 636 those servers. When additional framing is utilized, the usage will 637 specify how the client knows to apply it and what port to connect to. 638 For example, in the case of ICE connectivity checks, this information 639 is learned through out-of-band negotiation between client and server. 641 When STUN is run by itself over TLS-over-TCP, the 642 TLS_RSA_WITH_AES_128_CBC_SHA ciphersuite MUST be supported at a 643 minimum. Implementations MAY also support any other ciphersuite. 644 When it receives the TLS Certificate message, the client SHOULD 645 verify the certificate and inspect the site identified by the 646 certificate. If the certificate is invalid, revoked, or if it does 647 not identify the appropriate party, the client MUST NOT send the STUN 648 message or otherwise proceed with the STUN transaction. The client 649 MUST verify the identity of the server. To do that, it follows the 650 identification procedures defined in Section 3.1 of RFC 2818 651 [RFC2818]. Those procedures assume the client is dereferencing a 652 URI. For purposes of usage with this specification, the client 653 treats the domain name or IP address used in Section 8.1 as the host 654 portion of the URI that has been dereferenced. Alternatively, a 655 client MAY be configured with a set of domains or IP addresses that 656 are trusted; if a certificate is received that identifies one of 657 those domains or IP addresses, the client considers the identity of 658 the server to be verified. 660 When STUN is run multiplexed with other protocols over a TLS-over-TCP 661 connection, the mandatory ciphersuites and TLS handling procedures 662 operate as defined by those protocols. 664 Reliability of STUN over TCP and TLS-over-TCP is handled by TCP 665 itself, and there are no retransmissions at the STUN protocol level. 666 However, for a request/response transaction, if the client has not 667 received a response by Ti seconds after it sent the SYN to establish 668 the connection, it considers the transaction to have timed out. Ti 669 SHOULD be configurable and SHOULD have a default of 39.5s. This 670 value has been chosen to equalize the TCP and UDP timeouts for the 671 default initial RTO. 673 In addition, if the client is unable to establish the TCP connection, 674 or the TCP connection is reset or fails before a response is 675 received, any request/response transaction in progress is considered 676 to have failed 678 The client MAY send multiple transactions over a single TCP (or TLS- 679 over-TCP) connection, and it MAY send another request before 680 receiving a response to the previous. The client SHOULD keep the 681 connection open until it 683 o has no further STUN requests or indications to send over that 684 connection, and; 686 o has no plans to use any resources (such as a mapped address 687 (MAPPED-ADDRESS or XOR-MAPPED-ADDRESS) or relayed address 688 [I-D.ietf-behave-turn]) that were learned though STUN requests 689 sent over that connection, and; 691 o if multiplexing other application protocols over that port, has 692 finished using that other application, and; 694 o if using that learned port with a remote peer, has established 695 communications with that remote peer, as is required by some TCP 696 NAT traversal techniques (e.g., [I-D.ietf-mmusic-ice-tcp]). 698 At the server end, the server SHOULD keep the connection open, and 699 let the client close it, unless the server has determined that the 700 connection has timed out (for example, due to the client 701 disconnecting from the network). Bindings learned by the client will 702 remain valid in intervening NATs only while the connection remains 703 open. Only the client knows how long it needs the binding. The 704 server SHOULD NOT close a connection if a request was received over 705 that connection for which a response was not sent. A server MUST NOT 706 ever open a connection back towards the client in order to send a 707 response. Servers SHOULD follow best practices regarding connection 708 management in cases of overload. 710 7.3. Receiving a STUN Message 712 This section specifies the processing of a STUN message. The 713 processing specified here is for STUN messages as defined in this 714 specification; additional rules for backwards compatibility are 715 defined in in Section 12. Those additional procedures are optional, 716 and usages can elect to utilize them. First, a set of processing 717 operations are applied that are independent of the class. This is 718 followed by class-specific processing, described in the subsections 719 which follow. 721 When a STUN agent receives a STUN message, it first checks that the 722 message obeys the rules of Section 6. It checks that the first two 723 bits are 0, that the magic cookie field has the correct value, that 724 the message length is sensible, and that the method value is a 725 supported method. If the message-class is Success Response or Error 726 Response, the agent checks that the transaction ID matches a 727 transaction that is still in progress. If the FINGERPRINT extension 728 is being used, the agent checks that the FINGERPRINT attribute is 729 present and contains the correct value. If any errors are detected, 730 the message is silently discarded. In the case when STUN is being 731 multiplexed with another protocol, an error may indicate that this is 732 not really a STUN message; in this case, the agent should try to 733 parse the message as a different protocol. 735 The STUN agent then does any checks that are required by a 736 authentication mechanism that the usage has specified (see 737 Section 10. 739 Once the authentication checks are done, the STUN agent checks for 740 unknown attributes and known-but-unexpected attributes in the 741 message. Unknown comprehension-optional attributes MUST be ignored 742 by the agent. Known-but-unexpected attributes SHOULD be ignored by 743 the agent. Unknown comprehension-required attributes cause 744 processing that depends on the message-class and is described below. 746 At this point, further processing depends on the message class of the 747 request. 749 7.3.1. Processing a Request 751 If the request contains one or more unknown comprehension-required 752 attributes, the server replies with an error response with an error 753 code of 420 (Unknown Attribute), and includes an UNKNOWN-ATTRIBUTES 754 attribute in the response that lists the unknown comprehension- 755 required attributes. 757 The server then does any additional checking that the method or the 758 specific usage requires. If all the checks succeed, the server 759 formulates a success response as described below. 761 If the request uses UDP transport and is a retransmission of a 762 request for which the server has already generated a success response 763 within the last 40 seconds, the server MUST retransmit the same 764 success response. One way for a server to do this is to remember all 765 transaction IDs received over UDP and their corresponding responses 766 in the last 40 seconds. Another way is to reprocess the request and 767 recompute the response. The latter technique MUST only be applied to 768 requests which are idempotent (a request is considered idempotent 769 when the same request can be safely repeated without impacting the 770 overall state of the system) and result in the same success response 771 for the same request. The Binding method is considered to idempotent 772 in this way (even though certain rare network events could cause the 773 reflexive transport address value to change). Extensions to STUN 774 SHOULD state whether their request types have this property or not. 776 7.3.1.1. Forming a Success or Error Response 778 When forming the response (success or error), the server follows the 779 rules of section 6. The method of the response is the same as that 780 of the request, and the message class is either "Success Response" or 781 "Error Response". 783 For an error response, the server MUST add an ERROR-CODE attribute 784 containing the error code specified in the processing above. The 785 reason phrase is not fixed, but SHOULD be something suitable for the 786 error code. For certain errors, additional attributes are added to 787 the message. These attributes are spelled out in the description 788 where the error code is specified. For example, for an error code of 789 420 (Unknown Attribute), the server MUST include an UNKNOWN- 790 ATTRIBUTES attribute. Certain authentication errors also cause 791 attributes to be added (see Section 10). Extensions may define other 792 errors and/or additional attributes to add in error cases. 794 If the server authenticated the request using an authentication 795 mechanism, then the server SHOULD add the appropriate authentication 796 attributes to the response (see Section 10). 798 The server also adds any attributes required by the specific method 799 or usage. In addition, the server SHOULD add a SERVER attribute to 800 the message. 802 For the Binding method, no additional checking is required unless the 803 usage specifies otherwise. When forming the success response, the 804 server adds a XOR-MAPPED-ADDRESS attribute to the response, where the 805 contents of the attribute are the source transport address of the 806 request message. For UDP, this is the source IP address and source 807 UDP port of the request message. For TCP and TLS-over-TCP, this is 808 the source IP address and source TCP port of the TCP connection as 809 seen by the server. 811 7.3.1.2. Sending the Success or Error Response 813 The response (success or error) is sent over the same transport as 814 the request was received on. If the request was received over UDP, 815 the destination IP address and port of the response is the source IP 816 address and port of the received request message, and the source IP 817 address and port of the response is equal to the destination IP 818 address and port of the received request message. If the request was 819 received over TCP or TLS-over-TCP, the response is sent back on the 820 same TCP connection as the request was received on. 822 7.3.2. Processing an Indication 824 If the indication contains unknown comprehension-required attributes, 825 the indication is discarded and processing ceases. 827 The agent then does any additional checking that the method or the 828 specific usage requires. If all the checks succeed, the agent then 829 processes the indication. No response is generated for an 830 indication. 832 For the Binding method, no additional checking or processing is 833 required, unless the usage specifies otherwise. The mere receipt of 834 the message by the agent has refreshed the "bindings" in the 835 intervening NATs. 837 Since indications are not re-transmitted over UDP (unlike requests), 838 there is no need to handle re-transmissions of indications at the 839 sending agent. 841 7.3.3. Processing a Success Response 843 If the success response contains unknown comprehension-required 844 attributes, the response is discarded and the transaction is 845 considered to have failed. 847 The client then does any additional checking that the method or the 848 specific usage requires. If all the checks succeed, the client then 849 processes the success response. 851 For the Binding method, the client checks that the XOR-MAPPED-ADDRESS 852 attribute is present in the response. The client checks the address 853 family specified. If it is an unsupported address family, the 854 attribute SHOULD be ignored. If it is an unexpected but supported 855 address family (for example, the Binding transaction was sent over 856 IPv4, but the address family specified is IPv6), then the client MAY 857 accept and use the value. 859 7.3.4. Processing an Error Response 861 If the error response contains unknown comprehension-required 862 attributes, or if the error response does not contain an ERROR-CODE 863 attribute, then the transaction is simply considered to have failed. 865 The client then does any processing specified by the authentication 866 mechanism (see Section 10). This may result in a new transaction 867 attempt. 869 The processing at this point depends on the error-code, the method, 870 and the usage; the following are the default rules: 872 o If the error code is 300 through 399, the client SHOULD consider 873 the transaction as failed unless the ALTERNATE-SERVER extension is 874 being used. See Section 11. 876 o If the error code is 400 through 499, the client declares the 877 transaction failed; in the case of 420 (Unknown Attribute), the 878 response should contain a UNKNOWN-ATTRIBUTES attribute that gives 879 additional information. 881 o If the error code is 500 through 599, the client MAY resend the 882 request; clients that do so MUST limit the number of times they do 883 this. 885 Any other error code causes the client to consider the transaction 886 failed. 888 8. FINGERPRINT Mechanism 890 This section describes an optional mechanism for STUN that aids in 891 distinguishing STUN messages from packets of other protocols when the 892 two are multiplexed on the same transport address. This mechanism is 893 optional, and a STUN usage must describe if and when it is used. The 894 FINGERPRINT mechanism is not backwards compatible with RFC3489, and 895 cannot be used in environments where such compatibility is required. 897 In some usages, STUN messages are multiplexed on the same transport 898 address as other protocols, such as RTP. In order to apply the 899 processing described in Section 7, STUN messages must first be 900 separated from the application packets. Section 6 describes three 901 fixed fields in the STUN header that can be used for this purpose. 902 However, in some cases, these three fixed fields may not be 903 sufficient. 905 When the FINGERPRINT extension is used, an agent includes the 906 FINGERPRINT attribute in messages it sends to another agent. 907 Section 15.5 describes the placement and value of this attribute. 908 When the agent receives what it believes is a STUN message, then, in 909 addition to other basic checks, the agent also checks that the 910 message contains a FINGERPRINT attribute and that the attribute 911 contains the correct value. Section 7.3 describes when in the 912 overall processing of a STUN message the FINGERPRINT check is 913 performed. This additional check helps the agent detect messages of 914 other protocols that might otherwise seem to be STUN messages. 916 9. DNS Discovery of a Server 918 This section describes an optional procedure for STUN that allows a 919 client to use DNS to determine the IP address and port of a server. 920 A STUN usage must describe if and when this extension is used. To 921 use this procedure, the client must know a server's domain name and a 922 service name; the usage must also describe how the client obtains 923 these. Hard-coding the domain-name of the server into software is 924 NOT RECOMMENDED in case the domain name is lost or needs to change 925 for legal or other reasons. 927 When a client wishes to locate a STUN server in the public Internet 928 that accepts Binding Request/Response transactions, the SRV service 929 name is "stun". When it wishes to locate a STUN server which accepts 930 Binding Request/Response transactions over a TLS session, the SRV 931 service name is "stuns". STUN usages MAY define additional DNS SRV 932 service names. 934 The domain name is resolved to a transport address using the SRV 935 procedures specified in [RFC2782]. The DNS SRV service name is the 936 service name provided as input to this procedure. The protocol in 937 the SRV lookup is the transport protocol the client will run STUN 938 over: "udp" for UDP and "tcp" for TCP. Note that only "tcp" is 939 defined with "stuns" at this time. 941 The procedures of RFC 2782 are followed to determine the server to 942 contact. RFC 2782 spells out the details of how a set of SRV records 943 are sorted and then tried. However, RFC2782 only states that the 944 client should "try to connect to the (protocol, address, service)" 945 without giving any details on what happens in the event of failure. 946 When following these procedures, if the STUN transaction times out 947 without receipt of a response, the client SHOULD retry the request to 948 the next server in the ordered defined by RFC 2782. Such a retry is 949 only possible for request/response transmissions, since indication 950 transactions generate no response or timeout. 952 The default port for STUN requests is 3478, for both TCP and UDP. 953 Administrators of STUN servers SHOULD use this port in their SRV 954 records for UDP and TCP. In all cases, the port in DNS MUST reflect 955 the one the server is listening on. The default port for STUN over 956 TLS is XXXX [[NOTE TO RFC EDITOR: Replace with IANA registered port 957 number for stuns]]. Servers can run STUN over TLS on the same port 958 as STUN over TCP if the server software supports determining whether 959 the initial message is a TLS or STUN message. 961 If no SRV records were found, the client performs an A or AAAA record 962 lookup of the domain name. The result will be a list of IP 963 addresses, each of which can be contacted at the default port using 964 UDP or TCP, independent of the STUN usage. For usages that require 965 TLS, the client connects to one of the IP addresses using the default 966 STUN over TLS port. 968 10. Authentication and Message-Integrity Mechanisms 970 This section defines two mechanisms for STUN that a client and server 971 can use to provide authentication and message-integrity; these two 972 mechanisms are known as the short-term credential mechanism and the 973 long-term credential mechanism. These two mechanisms are optional, 974 and each usage must specify if and when these mechanisms are used. 975 Consequently, both clients and servers will know which mechanism (if 976 any) to follow based on knowledge of which usage applies. For 977 example, a STUN server on the public Internet supporting ICE would 978 have no authentication, whereas the STUN server functionality in an 979 agent supporting connectivity checks would utilize short term 980 credentials. An overview of these two mechanisms is given in 981 Section 3. 983 Each mechanism specifies the additional processing required to use 984 that mechanism, extending the processing specified in Section 7. The 985 additional processing occurs in three different places: when forming 986 a message; when receiving a message immediately after the basic 987 checks have been performed; and when doing the detailed processing of 988 error responses. 990 10.1. Short-Term Credential Mechanism 992 The short-term credential mechanism assumes that, prior to the STUN 993 transaction, the client and server have used some other protocol to 994 exchange a credential in the form of a username and password. This 995 credential is time-limited. The time-limit is defined by the usage. 996 As an example, in the ICE usage [I-D.ietf-mmusic-ice], the two 997 endpoints use out-of-band signaling to agree on a username and 998 password, and this username and password is applicable for the 999 duration of the media session. 1001 This credential is used to form a message integrity check in each 1002 request and in many responses. There is no challenge and response as 1003 in the long term mechanism; consequently, replay is prevented by 1004 virtue of the time-limited nature of the credential. 1006 10.1.1. Forming a Request or Indication 1008 For a request or indication message, the agent MUST include the 1009 USERNAME and MESSAGE-INTEGRITY attributes in the message. The HMAC 1010 for the MESSAGE-INTEGRITY attribute is computed as described in 1011 Section 15.4. Note that the password is never included in the 1012 request or indication. 1014 10.1.2. Receiving a Request or Indication 1016 After the agent has done the basic processing of a message, the agent 1017 performs the checks listed below in order specified: 1019 o If the message does not contain both a MESSAGE-INTEGRITY and a 1020 USERNAME attribute: 1022 * If the message is a request, the server MUST reject the request 1023 with an error response. This response MUST use an error code 1024 of 400 (Bad Request). 1026 * If the message is an indication, the agent MUST silently 1027 discard the indication. 1029 o If the USERNAME does not contain a username value currently valid 1030 within the server: 1032 * If the message is a request, the server MUST reject the request 1033 with an error response. This response MUST use an error code 1034 of 401 (Unauthorized). 1036 * If the message is an indication, the agent MUST silently 1037 discard the indication. 1039 o Using the password associated with the username, compute the value 1040 for the message-integrity as described in Section 15.4. If the 1041 resulting value does not match the contents of the MESSAGE- 1042 INTEGRITY attribute: 1044 * If the message is a request, the server MUST reject the request 1045 with an error response. This response MUST use an error code 1046 of 401 (Unauthorized). 1048 * If the message is an indication, the agent MUST silently 1049 discard the indication. 1051 If these checks pass, the agent continues to process the request or 1052 indication. Any response generated by a server MUST include the 1053 MESSAGE-INTEGRITY attribute, computed using the password utilized to 1054 authenticate the request. The response MUST NOT contain the USERNAME 1055 attribute. 1057 If any of the checks fail, a server MUST NOT include a MESSAGE- 1058 INTEGRITY or USERNAME attribute in the error response. This is 1059 because, in these failure cases, the server cannot determine the 1060 shared secret necessary to compute MESSAGE-INTEGRITY. 1062 10.1.3. Receiving a Response 1064 The client looks for the MESSAGE-INTEGRITY attribute in the response. 1065 If present, the client computes the message integrity over the 1066 response as defined in Section 15.4, using the same password it 1067 utilized for the request. If the resulting value matches the 1068 contents of the MESSAGE-INTEGRITY attribute, the response is 1069 considered authenticated. If the value does not match, or if 1070 MESSAGE-INTEGRITY was absent, the response MUST be discarded, as if 1071 it was never received. This means that retransmits, if applicable, 1072 will continue. 1074 10.2. Long-term Credential Mechanism 1076 The long-term credential mechanism relies on a long term credential, 1077 in the form of a username and password, that are shared between 1078 client and server. The credential is considered long-term since it 1079 is assumed that it is provisioned for a user, and remains in effect 1080 until the user is no longer a subscriber of the system, or is 1081 changed. This is basically a traditional "log-in" username and 1082 password given to users. 1084 Because these usernames and passwords are expected to be valid for 1085 extended periods of time, replay prevention is provided in the form 1086 of a digest challenge. In this mechanism, the client initially sends 1087 a request, without offering any credentials or any integrity checks. 1088 The server rejects this request, providing the user a realm (used to 1089 guide the user or agent in selection of a username and password) and 1090 a nonce. The nonce provides the replay protection. It is a cookie, 1091 selected by the server, and encoded in such a way as to indicate a 1092 duration of validity or client identity from which it is valid. The 1093 client retries the request, this time including its username, the 1094 realm, and echoing the nonce provided by the server. The client also 1095 includes a message-integrity, which provides an HMAC over the entire 1096 request, including the nonce. The server validates the nonce, and 1097 checks the message-integrity. If they match, the request is 1098 authenticated. If the nonce is no longer valid, it is considered 1099 "stale", and the server rejects the request, providing a new nonce. 1101 In subsequent requests to the same server, the client reuses the 1102 nonce, username, realm and password it used previously. In this way, 1103 subsequent requests are not rejected until the nonce becomes invalid 1104 by the server, in which case the rejection provides a new nonce to 1105 the client. 1107 Note that the long-term credential mechanism cannot be used to 1108 protect indications, since indications cannot be challenged. Usages 1109 utilizing indications must either use a short-term credential, or 1110 omit authentication and message integrity for them. 1112 Since the long-term credential mechanism is susceptible to offline 1113 dictionary attacks, deployments SHOULD utilize strong passwords. 1115 10.2.1. Forming a Request 1117 There are two cases when forming a request. In the first case, this 1118 is the first request from the client to the server (as identified by 1119 its IP address and port). In the second case, the client is 1120 submitting a subsequent request once a previous request/response 1121 transaction has completed successfully. Forming a request as a 1122 consequence of a 401 or 438 error response is covered in 1123 Section 10.2.3 and is not considered a "subsequent request" and thus 1124 does not utilize the rules described in Section 10.2.1.2. 1126 10.2.1.1. First Request 1128 If the client has not completed a successful request/response 1129 transaction with the server (as identified by hostname, if the DNS 1130 procedures of Section 9 are used, else IP address if not), it SHOULD 1131 omit the USERNAME, MESSAGE-INTEGRITY, REALM, and NONCE attributes. 1132 In other words, the very first request is sent as if there were no 1133 authentication or message integrity applied. The exception to this 1134 rule are requests sent to another server as a consequence of the 1135 ALTERNATE-SERVER mechanism described in Section 11. Those requests 1136 do include the USERNAME, REALM and NONCE from the original request, 1137 along with a newly computed MESSAGE-INTEGRITY based on them. 1139 10.2.1.2. Subsequent Requests 1141 Once a request/response transaction has completed successfully, the 1142 client will have been been presented a realm and nonce by the server, 1143 and selected a username and password with which it authenticated. 1144 The client SHOULD cache the username, password, realm, and nonce for 1145 subsequent communications with the server. When the client sends a 1146 subsequent request, it SHOULD include the USERNAME, REALM, and NONCE 1147 attributes with these cached values. It SHOULD include a MESSAGE- 1148 INTEGRITY attribute, computed as described in Section 15.4 using the 1149 cached password. 1151 10.2.2. Receiving a Request 1153 After the server has done the basic processing of a request, it 1154 performs the checks listed below in the order specified: 1156 o If the message does not contain a MESSAGE-INTEGRITY attribute, the 1157 server MUST generate an error response with an error code of 401 1158 (Unauthorized). This response MUST include a REALM value. It is 1159 RECOMMENDED that the REALM value be the domain name of the 1160 provider of the STUN server. The response MUST include a NONCE, 1161 selected by the server. The response SHOULD NOT contain a 1162 USERNAME or MESSAGE-INTEGRITY attribute. 1164 o If the message contains a MESSAGE-INTEGRITY attribute, but is 1165 missing the USERNAME, REALM or NONCE attributes, the server MUST 1166 generate an error response with an error code of 400 (Bad 1167 Request). This response SHOULD NOT include a USERNAME, NONCE, 1168 REALM or MESSAGE-INTEGRITY attribute. 1170 o If the NONCE is no longer valid, the server MUST generate an error 1171 response with an error code of 438 (Stale Nonce). This response 1172 MUST include a NONCE and REALM attribute and SHOULD NOT incude the 1173 USERNAME or MESSAGE-INTEGRITY attribute. Servers can invalidate 1174 nonces in order to provide additional security. See Section 4.3 1175 of [RFC2617] for guidelines. 1177 o If the username in the USERNAME attribute is not valid, the server 1178 MUST generate an error response with an error code of 401 1179 (Unauthorized). This response MUST include a REALM value. It is 1180 RECOMMENDED that the REALM value be the domain name of the 1181 provider of the STUN server. The response MUST include a NONCE, 1182 selected by the server. The response SHOULD NOT contain a 1183 USERNAME or MESSAGE-INTEGRITY attribute. 1185 o Using the password associated with the username in the USERNAME 1186 attribute, compute the value for the message-integrity as 1187 described in Section 15.4. If the resulting value does not match 1188 the contents of the MESSAGE-INTEGRITY attribute, the server MUST 1189 reject the request with an error response. This response MUST use 1190 an error code of 401 (Unauthorized). It MUST include a REALM and 1191 NONCE attribute and SHOULD NOT include the USERNAME or MESSAGE- 1192 INTEGRITY attribute. 1194 If these checks pass, the server continues to process the request. 1195 Any response generated by the server (excepting the cases described 1196 above) MUST include the MESSAGE-INTEGRITY attribute, computed using 1197 the username and password utilized to authenticate the request. The 1198 REALM, NONCE, and USERNAME attributes SHOULD NOT be included. 1200 10.2.3. Receiving a Response 1202 If the response is an error response, with an error code of 401 1203 (Unauthorized), the client SHOULD retry the request with a new 1204 transaction. This request MUST contain a USERNAME, determined by the 1205 client as the appropriate username for the REALM from the error 1206 response. The request MUST contain the REALM, copied from the error 1207 response. The request MUST contain the NONCE, copied from the error 1208 response. The request MUST contain the MESSAGE-INTEGRITY attribute, 1209 computed using the password associated with the username in the 1210 USERNAME attribute. The client MUST NOT perform this retry if it is 1211 not changing the USERNAME or REALM or its associated password, from 1212 the previous attempt. 1214 If the response is an error response with an error code of 438 (Stale 1215 Nonce), the client MUST retry the request, using the new NONCE 1216 supplied in the 438 (Stale Nonce) response. This retry MUST also 1217 include the USERNAME, REALM and MESSAGE-INTEGRITY. 1219 The client looks for the MESSAGE-INTEGRITY attribute in the response 1220 (either success or failure). If present, the client computes the 1221 message integrity over the response as defined in Section 15.4, using 1222 the same password it utilized for the request. If the resulting 1223 value matches the contents of the MESSAGE-INTEGRITY attribute, the 1224 response is considered authenticated. If the value does not match, 1225 or if MESSAGE-INTEGRITY was absent, the response MUST be discarded, 1226 as if it was never received. This means that retransmits, if 1227 applicable, will continue. 1229 11. ALTERNATE-SERVER Mechanism 1231 This section describes a mechanism in STUN that allows a server to 1232 redirect a client to another server. This extension is optional, and 1233 a usage must define if and when this extension is used. To prevent 1234 denial-of-service attacks, this extension MUST only be used in 1235 situations where the client and server are using an authentication 1236 and message-integrity mechanism. 1238 A server using this extension redirects a client to another server by 1239 replying to a request message with an error response message with an 1240 error code of 300 (Try Alternate). The server MUST include a 1241 ALTERNATE-SERVER attribute in the error response. The error response 1242 message MUST be authenticated, which in practice means the request 1243 message must have passed the authentication checks. 1245 A client using this extension handles a 300 (Try Alternate) error 1246 code as follows. If the error response has passed the authentication 1247 checks, then the client looks for a ALTERNATE-SERVER attribute in the 1248 error response. If one is found, then the client considers the 1249 current transaction as failed, and re-attempts the request with the 1250 server specified in the attribute, using the same transport protocol 1251 used for the previous request. The client SHOULD reuse any 1252 authentication credentials from the old request in the new 1253 transaction. If the server has been redirected to a server on which 1254 it has already tried this request within the last five minutes, it 1255 MUST ignore the redirection and consider the transaction to have 1256 failed. This prevents infinite ping-ponging between servers in case 1257 of redirection loops. 1259 12. Backwards Compatibility with RFC 3489 1261 This section defines procedures that allow a degree of backwards 1262 compatible with the original protocol defined in RFC 3489 [RFC3489]. 1263 This mechanism is optional, meant to be utilized only in cases where 1264 a new client can connect to an old server, or vice-a-versa. A usage 1265 must define if and when this procedure is used. 1267 Section 19 lists all the changes between this specification and RFC 1268 3489 [RFC3489]. However, not all of these differences are important, 1269 because "classic STUN" was only used in a few specific ways. For the 1270 purposes of this extension, the important changes are the following. 1271 In RFC 3489: 1273 o UDP was the only supported transport; 1275 o The field that is now the Magic Cookie field was a part of the 1276 transaction id field, and transaction ids were 128 bits long; 1278 o The XOR-MAPPED-ADDRESS attribute did not exist, and the Binding 1279 method used the MAPPED-ADDRESS attribute instead; 1281 o There were three comprehension-required attributes, RESPONSE- 1282 ADDRESS, CHANGE-REQUEST, and CHANGED-ADDRESS that have been 1283 removed from this specification; 1285 * These attributes are now part of the NAT Behavior Discovery 1286 usage. [I-D.ietf-behave-nat-behavior-discovery] 1288 12.1. Changes to Client Processing 1290 A client that wants to interoperate with a [RFC3489] server SHOULD 1291 send a request message that uses the Binding method, contains no 1292 attributes, and uses UDP as the transport protocol to the server. If 1293 successful, the success response received from the server will 1294 contain a MAPPED-ADDRESS attribute rather than an XOR-MAPPED-ADDRESS 1295 attribute. A client seeking to interoperate with an older server 1296 MUST be prepared to receive either. Furthermore, the client MUST 1297 ignore any Reserved comprehension-required attributes which might 1298 appear in the response. Of the Reserved attributes in in 1299 Section 18.2, 0x0002,0x0004,0x0005 and 0x000B may appear in Binding 1300 Responses from a server compliant to RFC 3489. Other than this 1301 change, the processing of the response is identical to the procedures 1302 described above. 1304 12.2. Changes to Server Processing 1306 A STUN server can detect when a given Binding Request message was 1307 sent from an RFC 3489 [RFC3489] client by the absence of the correct 1308 value in the Magic Cookie field. When the server detects an RFC 3489 1309 client, it SHOULD copy the value seen in the Magic Cookie field in 1310 the Binding Request to the Magic Cookie field in the Binding Response 1311 message, and insert a MAPPED-ADDRESS attribute instead of an XOR- 1312 MAPPED-ADDRESS attribute. 1314 The client might, in rare situations, include either the RESPONSE- 1315 ADDRESS or CHANGE-REQUEST attributes. In these situations, the 1316 server will view these as unknown comprehension-required attributes 1317 and reply with an error response. Since the mechanisms utilizing 1318 those attributes are no longer supported, this behavior is 1319 acceptable. 1321 The RFC 3489 version of STUN lacks both the Magic Cookie and the 1322 FINGERPRINT attribute that allows for a very high probablility of 1323 correctly identifying STUN messages when multiplexed with other 1324 protocols. Therefore, STUN implementations that are backwards 1325 compatible with RFC 3489 SHOULD NOT be used in cases where STUN will 1326 be multiplexed with another protocol. However, that should not be an 1327 issues as such multiplexing was not available in RFC 3489. 1329 13. Basic Server Behavior 1331 This section defines the behavior of a basic, standalone STUN server. 1332 A basic STUN server provides clients with server reflexive transport 1333 addresses by receiving and replying to STUN Binding Requests. 1335 The STUN server MUST support the Binding method. It SHOULD NOT 1336 utilize the short term or long term credential mechanism. This is 1337 because the work involved in authenticating the request is more than 1338 the work in simply processing it. It SHOULD NOT utilize the 1339 ALTERNATE-SERVER mechanism for the same reason. It MUST support UDP 1340 and TCP. It MAY support STUN over TCP/TLS, however TLS provides 1341 minimal security benefits in this basic mode of operation. It MAY 1342 utilize the FINGERPRINT mechanism but MUST NOT require it. Since the 1343 standalove server only runs STUN, FINGERPRINT provides no benefit. 1344 Requiring it would break compatibility with RFC 3489, and such 1345 compatibility is desirable in a standalone server. Standalone STUN 1346 servers SHOULD support backwards compatibility with [RFC3489] 1347 clients, as described in Section 12. 1349 It is RECOMMENDED that administrators of STUN servers provide DNS 1350 entries for those servers as described in Section 9. 1352 A basic STUN server is not a solution for NAT traversal by itself. 1353 However, it can be utilized as part of a solution through STUN 1354 usages. This is discussed further in Section 14. 1356 14. STUN Usages 1358 STUN by itself is not a solution to the NAT traversal problem. 1359 Rather, STUN defines a tool that can be used inside a larger 1360 solution. The term "STUN Usage" is used for any solution that uses 1361 STUN as a component. 1363 At the time of writing, three STUN usages are defined: Interactive 1364 Connectivity Establishment (ICE) [I-D.ietf-mmusic-ice], Client- 1365 initiated connections for SIP [I-D.ietf-sip-outbound], and NAT 1366 Behavior Discovery [I-D.ietf-behave-nat-behavior-discovery]. Other 1367 STUN usages may be defined in the future. 1369 A STUN usage defines how STUN is actually utilized - when to send 1370 requests, what to do with the responses, and which optional 1371 procedures defined here (or in an extension to STUN) are to be used. 1372 A usage would also define: 1374 o Which STUN methods are used; 1376 o What authentication and message integrity mechanisms are used; 1378 o The considerations around manual vs. automatic key derivation for 1379 the integrity mechanism, as discussed in [RFC4107]; 1381 o What mechanisms are used to distinguish STUN messages from other 1382 messages. When STUN is run over TCP, a framing mechanism may be 1383 required; 1385 o How a STUN client determines the IP address and port of the STUN 1386 server; 1388 o Whether backwards compatibility to RFC 3489 is required; 1390 o What optional attributes defined here (such as FINGERPRINT and 1391 ALTERNATE-SERVER) or in other extensions are required. 1393 In addition, any STUN usage must consider the security implications 1394 of using STUN in that usage. A number of attacks against STUN are 1395 known (see the Security Considerations section in this document) and 1396 any usage must consider how these attacks can be thwarted or 1397 mitigated. 1399 Finally, a usage must consider whether its usage of STUN is an 1400 example of the Unilateral Self-Address Fixing approach to NAT 1401 traversal, and if so, address the questions raised in RFC 3424. 1402 [RFC3424] 1404 15. STUN Attributes 1406 After the STUN header are zero or more attributes. Each attribute 1407 MUST be TLV encoded, with a 16 bit type, 16 bit length, and value. 1408 Each STUN attribute MUST end on a 32 bit boundary. As mentioned 1409 above, all fields in an attribute are transmitted most significant 1410 bit first. 1412 0 1 2 3 1413 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1414 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1415 | Type | Length | 1416 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1417 | Value (variable) .... 1418 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1420 Figure 4: Format of STUN Attributes 1422 The value in the Length field MUST contain the length of the Value 1423 part of the attribute, prior to padding, measured in bytes. Since 1424 STUN aligns attributes on 32 bit boundaries, attributes whose content 1425 is not a multiple of 4 bytes are padded with 1, 2 or 3 bytes of 1426 padding so that its value contains a multiple of 4 bytes. The 1427 padding bits are ignored, and may be any value. 1429 Any attribute type MAY appear more than once in a STUN message. 1430 Unless specified otherwise, the order of appearance is significant: 1431 only the first occurance needs to be processed by a receiver, and any 1432 duplicates MAY be ignored by a receiver. 1434 To allow future revisions of this specification to add new attributes 1435 if needed, the attribute space is divided into two ranges. 1436 Attributes with type values between 0x0000 and 0x7FFF are 1437 comprehension-required attributes, which means that the STUN agent 1438 cannot successfully process the message unless it understands the 1439 attribute. Attributes with type values between 0x8000 and 0xFFFF are 1440 comprehension-optional attributes, which means that those attributes 1441 can be ignored by the STUN agent if it does not understand them. 1443 The set of STUN attribute types is maintained by IANA. The initial 1444 set defined by this specification is found in Section 18.2. 1446 The rest of this section describes the format of the various 1447 attributes defined in this specification. 1449 15.1. MAPPED-ADDRESS 1451 The MAPPED-ADDRESS attribute indicates a reflexive transport address 1452 of the client. It consists of an eight bit address family, and a 1453 sixteen bit port, followed by a fixed length value representing the 1454 IP address. If the address family is IPv4, the address MUST be 32 1455 bits. If the address family is IPv6, the address MUST be 128 bits. 1456 All fields must be in network byte order. 1458 The format of the MAPPED-ADDRESS attribute is: 1460 0 1 2 3 1461 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1462 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1463 |0 0 0 0 0 0 0 0| Family | Port | 1464 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1465 | | 1466 | Address (32 bits or 128 bits) | 1467 | | 1468 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1470 Figure 5: Format of MAPPED-ADDRESS attribute 1472 The address family can take on the following values: 1474 0x01:IPv4 1475 0x02:IPv6 1477 The first 8 bits of the MAPPED-ADDRESS MUST be set to 0 and MUST be 1478 ignored by receivers. These bits are present for aligning parameters 1479 on natural 32 bit boundaries. 1481 This attribute is used only by servers for achieving backwards 1482 compatibility with RFC 3489 [RFC3489] clients. 1484 15.2. XOR-MAPPED-ADDRESS 1486 The XOR-MAPPED-ADDRESS attribute is identical to the MAPPED-ADDRESS 1487 attribute, except that the reflexive transport address is obfuscated 1488 through the XOR function. 1490 The format of the XOR-MAPPED-ADDRESS is: 1492 0 1 2 3 1493 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1494 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1495 |x x x x x x x x| Family | X-Port | 1496 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1497 | X-Address (Variable) 1498 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1500 Figure 7: Format of XOR-MAPPED-ADDRESS Attribute 1502 The Family represents the IP address family, and is encoded 1503 identically to the Family in MAPPED-ADDRESS. 1505 X-Port is computed by taking the mapped port in host byte order, 1506 XOR'ing it with the most significant 16 bits of the magic cookie, and 1507 then the converting the result to network byte order. If the IP 1508 address family is IPv4, X-Address is computed by taking the mapped IP 1509 address in host byte order, XOR'ing it with the magic cookie, and 1510 converting the result to network byte order. If the IP address 1511 family is IPv6, X-Address is computed by taking the mapped IP address 1512 in host byte order, XOR'ing it with the concatenation of the magic 1513 cookie and the 96-bit transaction ID, and converting the result to 1514 network byte order. 1516 The rules for encoding and processing the first 8 bits of the 1517 attribute's value, the rules for handling multiple occurrences of the 1518 attribute, and the rules for processing addresses families are the 1519 same as for MAPPED-ADDRESS. 1521 NOTE: XOR-MAPPED-ADDRESS and MAPPED-ADDRESS differ only in their 1522 encoding of the transport address. The former encodes the transport 1523 address by exclusive-or'ing it with the magic cookie. The latter 1524 encodes it directly in binary. RFC 3489 originally specified only 1525 MAPPED-ADDRESS. However, deployment experience found that some NATs 1526 rewrite the 32-bit binary payloads containing the NAT's public IP 1527 address, such as STUN's MAPPED-ADDRESS attribute, in the well-meaning 1528 but misguided attempt at providing a generic ALG function. Such 1529 behavior interferes with the operation of STUN and also causes 1530 failure of STUN's message integrity checking. 1532 15.3. USERNAME 1534 The USERNAME attribute is used for message integrity. It identifies 1535 the username and password combination used in the message integrity 1536 check. 1538 The value of USERNAME is a variable length value. It MUST contain a 1539 UTF-8 [RFC3629] encoded sequence of less than 513 bytes, and MUST 1540 have been processed using SASLPrep [RFC4013]. 1542 15.4. MESSAGE-INTEGRITY 1544 The MESSAGE-INTEGRITY attribute contains an HMAC-SHA1 [RFC2104] of 1545 the STUN message. The MESSAGE-INTEGRITY attribute can be present in 1546 any STUN message type. Since it uses the SHA1 hash, the HMAC will be 1547 20 bytes. The text used as input to HMAC is the STUN message, 1548 including the header, up to and including the attribute preceding the 1549 MESSAGE-INTEGRITY attribute. With the exception of the FINGERPRINT 1550 attribute, which appears after MESSAGE-INTEGRITY, agents MUST ignore 1551 all other attributes that follow MESSAGE-INTEGRITY. 1553 The key for the HMAC depends on whether long term or short term 1554 credentials are in use. For long term credentials, the key is 16 1555 bytes: 1557 key = MD5(username ":" realm ":" SASLPrep(password)) 1559 That is, the 16 byte key is formed by taking the MD5 hash of the 1560 result of concatenating the following five fields: (1) The username, 1561 with any quotes and trailing nulls removed, as taken from the 1562 USERNAME attribute (in which case SASLPrep has already been applied) 1563 (2) A single colon, (3) The realm, with any quotes and trailing nulls 1564 removed, (4) A single colon, and (5) the password, with any trailing 1565 nulls removed and after processing using SASLPrep. For example, if 1566 the username was 'user', the realm was 'realm', and the password was 1567 'pass', then the 16-byte HMAC key would be the result of performing 1568 an MD5 hash on the string 'user:realm:pass', the resulting hash being 1569 0x8493fbc53ba582fb4c044c456bdc40eb. 1571 For short term credentials: 1573 key = SASLPrep(password) 1575 Where MD5 is defined in RFC 1321 [RFC1321] and SASLPrep() is defined 1576 in [RFC4013]. 1578 The structure of the key when used with long term credentials 1579 facilitates deployment in systems that also utilize SIP. Typically, 1580 SIP systems utilizing SIP's digest authentication mechanism do not 1581 actually store the password in the database. Rather, they store a 1582 value called H(A1), which is equal to the key defined above. 1584 Based on the rules above, the hash includes the length field from the 1585 STUN message header. Prior to performing the hash, the MESSAGE- 1586 INTEGRITY attribute MUST be inserted into the message (with dummy 1587 content). The length MUST then be set to point to the length of the 1588 message up to, and including, the MESSAGE-INTEGRITY attribute itself, 1589 but excluding any attributes after it. Once the computation is 1590 performed, the value of the MESSAGE-INTEGRITY attribute can be filled 1591 in, and the value of the length in the STUN header can be set to its 1592 correct value - the length of the entire message. Similarly, when 1593 validating the MESSAGE-INTEGRITY, the length field should be adjusted 1594 to point to the end of the MESSAGE-INTEGRITY attribute prior to 1595 calculating the HMAC. Such adjustment is necessary when attributes, 1596 such as FINGERPRINT, appear after MESSAGE-INTEGRITY. 1598 15.5. FINGERPRINT 1600 The FINGERPRINT attribute MAY be present in all STUN messages. The 1601 value of the attribute is computed as the CRC-32 of the STUN message 1602 up to (but excluding) the FINGERPRINT attribute itself, xor-d with 1603 the 32 bit value 0x5354554e (the XOR helps in cases where an 1604 application packet is also using CRC-32 in it). The 32 bit CRC is 1605 the one defined in ITU V.42 [ITU.V42.2002], which has a generator 1606 polynomial of x32+x26+x23+x22+x16+x12+x11+x10+x8+x7+x5+x4+x2+x+1. 1607 When present, the FINGERPRINT attribute MUST be the last attribute in 1608 the message, and thus will appear after MESSAGE-INTEGRITY. 1610 The FINGERPRINT attribute can aid in distinguishing STUN packets from 1611 packets of other protocols. See Section 8. 1613 As with MESSAGE-INTEGRITY, the CRC used in the FINGERPRINT attribute 1614 covers the length field from the STUN message header. Therefore, 1615 this value must be correct, and include the CRC attribute as part of 1616 the message length, prior to computation of the CRC. When using the 1617 FINGERPRINT attribute in a message, the attribute is first placed 1618 into the message with a dummy value, then the CRC is computed, and 1619 then the value of the attribute is updated. If the MESSAGE-INTEGRITY 1620 attribute is also present, then it must be present with the correct 1621 message-integrity value before the CRC is computed, since the CRC is 1622 done over the value of the MESSAGE-INTEGRITY attribute as well. 1624 15.6. ERROR-CODE 1626 The ERROR-CODE attribute is used in Error Response messages. It 1627 contains a numeric error code value in the range of 300 to 699 plus a 1628 textual reason phrase encoded in UTF-8 [RFC3629], and is consistent 1629 in its code assignments and semantics with SIP [RFC3261] and HTTP 1630 [RFC2616]. The reason phrase is meant for user consumption, and can 1631 be anything appropriate for the error code. Recommended reason 1632 phrases for the defined error codes are presented below. The reason 1633 phrase MUST be a UTF-8 [RFC3629] encoded sequence of less than 128 1634 characters (which can be as long as 763 bytes). 1636 0 1 2 3 1637 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1638 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1639 | Reserved, should be 0 |Class| Number | 1640 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1641 | Reason Phrase (variable) .. 1642 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1644 Figure 10: ERROR-CODE Attribute 1646 To facilitate processing, the class of the error code (the hundreds 1647 digit) is encoded separately from the rest of the code, as shown in 1648 Figure 10. 1650 The Reserved bits SHOULD be 0, and are for alignment on 32-bit 1651 boundaries. Receivers MUST ignore these bits. The Class represents 1652 the hundreds digit of the error code. The value MUST be between 3 1653 and 6. The number represents the error code modulo 100, and its 1654 value MUST be between 0 and 99. 1656 The following error codes, along with their recommended reason 1657 phrases are defined: 1659 300 Try Alternate: The client should contact an alternate server for 1660 this request. This error response MUST only be sent if the 1661 request included a USERNAME attribute and a valid MESSAGE- 1662 INTEGRITY attribute; otherwise it MUST NOT be sent and error 1663 code 400 (Bad Request) is suggested. This error response MUST 1664 be protected with the MESSAGE-INTEGRITY attribute, and receivers 1665 MUST validate the MESSAGE-INTEGRITY of this response before 1666 redirecting themselves to an alternate server. 1668 Note: failure to generate and validate message-integrity 1669 for a 300 response allows an on-path attacker to falsify a 1670 300 response thus causing subsequent STUN messages to be 1671 sent to a victim. 1673 400 Bad Request: The request was malformed. The client SHOULD NOT 1674 retry the request without modification from the previous 1675 attempt. The server may not be able to generate a valid 1676 MESSAGE-INTEGRITY for this error, so the client MUST NOT expect 1677 a valid MESSAGE-INTEGRITY attribute on this response. 1679 401 Unauthorized: The request did not contain the correct 1680 credentials to proceed. The client should retry the request 1681 with proper credentials. 1683 420 Unknown Attribute: The server received STUN packet containing a 1684 comprehension-required attribute which it did not understand. 1685 The server MUST put this unknown attribute in the UNKNOWN- 1686 ATTRIBUTE attribute of its error response. 1688 438 Stale Nonce: The NONCE used by the client was no longer valid. 1689 The client should retry, using the NONCE provided in the 1690 response. 1692 500 Server Error: The server has suffered a temporary error. The 1693 client should try again. 1695 15.7. REALM 1697 The REALM attribute may be present in requests and responses. It 1698 contains text which meets the grammar for "realm-value" as described 1699 in RFC 3261 [RFC3261] but without the double quotes and their 1700 surrounding whitespace. That is, it is an unquoted realm-value (and 1701 is therefore a sequence of qdtext or quoted-pair). It MUST be a 1702 UTF-8 [RFC3629] encoded sequence of less than 128 characters (which 1703 can be as long as 763 bytes), and MUST have been processed using 1704 SASLPrep [RFC4013]. 1706 Presence of the REALM attribute in a request indicates that long-term 1707 credentials are being used for authentication. Presence in certain 1708 error responses indicates that the server wishes the client to use a 1709 long-term credential for authentication. 1711 15.8. NONCE 1713 The NONCE attribute may be present in requests and responses. It 1714 contains a sequence of qdtext or quoted-pair, which are defined in 1715 RFC 3261 [RFC3261]. Note that this means that the NONCE attribute 1716 will not contain actual quote characters. See RFC 2617 [RFC2617], 1717 Section 4.3, for guidance on selection of nonce values in a server. 1718 It MUST be less than 128 characters (which can be as long as 763 1719 bytes). 1721 15.9. UNKNOWN-ATTRIBUTES 1723 The UNKNOWN-ATTRIBUTES attribute is present only in an error response 1724 when the response code in the ERROR-CODE attribute is 420. 1726 The attribute contains a list of 16 bit values, each of which 1727 represents an attribute type that was not understood by the server. 1729 0 1 2 3 1730 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1731 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1732 | Attribute 1 Type | Attribute 2 Type | 1733 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1734 | Attribute 3 Type | Attribute 4 Type ... 1735 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1737 Figure 11: Format of UNKNOWN-ATTRIBUTES attribute 1739 Note: In [RFC3489], this field was padded to 32 by duplicating the 1740 last attribute. In this version of the specification, the normal 1741 padding rules for attributes are used instead. 1743 15.10. SERVER 1745 The server attribute contains a textual description of the software 1746 being used by the server. This SHOULD include manufacturer and 1747 version number. The attribute has no impact on operation of the 1748 protocol, and serves only as a tool for diagnostic and debugging 1749 purposes. The value of SERVER is variable length. It MUST be a 1750 UTF-8 [RFC3629] encoded sequence of less than 128 characters (which 1751 can be as long as 763 bytes). 1753 15.11. CLIENT 1755 The client attribute contains a textual description of the software 1756 being used by the client. This SHOULD include manufacturer and 1757 version number. The attribute has no impact on operation of the 1758 protocol, and serves only as a tool for diagnostic and debugging 1759 purposes. The value of CLIENT is variable length. It MUST be a 1760 UTF-8 [RFC3629] encoded sequence of less than 128 characters (which 1761 can be as long as 763 bytes). 1763 15.12. ALTERNATE-SERVER 1765 The alternate server represents an alternate transport address 1766 identifying a different STUN server which the STUN client should try. 1768 It is encoded in the same way as MAPPED-ADDRESS, and thus refers to a 1769 single server by IP address. The IP address family MUST be identical 1770 to that of the source IP address of the request. 1772 This attribute MUST only appear in an error response that contains a 1773 MESSAGE-INTEGRITY attribute. This prevents it from being used in 1774 denial-of-service attacks. 1776 16. Security Considerations 1778 16.1. Attacks against the Protocol 1780 16.1.1. Outside Attacks 1782 An attacker can try to modify STUN messages in transit, in order to 1783 cause a failure in STUN operation. These attacks are detected for 1784 both requests and responses through the message integrity mechanism, 1785 using either a short term or long term credential. Of course, once 1786 detected, the manipulated packets will be dropped, causing the STUN 1787 transaction to effectively fail. This attack is possible only by an 1788 on-path attacker. 1790 An attacker that can observe, but not modify STUN messages in-transit 1791 (for example, an attacker present on a shared access medium, such as 1792 Wi-Fi), can see a STUN request, and then immediately send a STUN 1793 response, typically an error response, in order to disrupt STUN 1794 processing. This attack is also prevented for messages that utilize 1795 MESSAGE-INTEGRITY. However, some error responses, those related to 1796 authentication in particular, cannot be protected by MESSAGE- 1797 INTEGRITY. When STUN itself is run over a secure transport protocol 1798 (e.g., TLS), these attacks are completely mitigated. 1800 Depending on the STUN usage, these attacks may be of minimal 1801 consequence and thus do not require message integrity to mitigate. 1802 For example, when STUN is used to a basic STUN server to discover a 1803 server reflexive candidate for usage with ICE, authentication and 1804 message integrity are not required since these attacks are detected 1805 during the connectivity check phase. The connectivity checks 1806 themselves, however, require protection for proper operation of ICE 1807 overall. As described in Section 14, STUN usages describe when 1808 authentication and message integrity are needed. 1810 Since STUN uses the HMAC of a shared secret for authentication and 1811 integrity protection, it is subject to offline dictionary attacks. 1812 When authentication is utilized, it SHOULD be with a strong password 1813 that is not readily subject to offline dictionary attacks. 1814 Protection of the channel itself, using TLS, mitigates these attacks. 1815 However, STUN is most often run over UDP and in those cases, strong 1816 passwords are the only way to protect against these attacks. 1818 16.1.2. Inside Attacks 1820 A rogue client may try to launch a DoS attack against a server by 1821 sending it a large number of STUN requests. Fortunately, STUN 1822 requests can be processed statelessly by a server, making such 1823 attacks hard to launch. 1825 A rogue client may use a STUN server as a reflector, sending it 1826 requests with a falsified source IP address and port. In such a 1827 case, the response would be delivered to that source IP and port. 1828 There is no amplification of the number of packets with this attack 1829 (the STUN server sends one packet for each packet sent by the 1830 client), though there is a small increase in the amount of data, 1831 since STUN responses are typically larger than requests. This attack 1832 is mitigated by ingress source address filtering. 1834 Revealing the specific software version of the server or client 1835 through the SERVER and CLIENT attributes might allow them to become 1836 more vulnerable to attacks against software that is known to contain 1837 security holes. Implementers SHOULD make the CLIENT and SERVER 1838 attributes a configurable option. 1840 16.2. Attacks Affecting the Usage 1842 This section lists attacks that might be launched against a usage of 1843 STUN. Each STUN usage must consider whether these attacks are 1844 applicable to it, and if so, discuss counter-measures. 1846 Most of the attacks in this section revolve around an attacker 1847 modifying the reflexive address learned by a STUN client through a 1848 Binding Request/Binding Response transaction. Since the usage of the 1849 reflexive address is a function of the usage, the applicability and 1850 remediation of these attacks is usage-specific. In common 1851 situations, modification of the reflexive address by an on-path 1852 attacker is easy to do. Consider, for example, the common situation 1853 where STUN is run directly over UDP. In this case, an on-path 1854 attacker can modify the source IP address of the Binding Request 1855 before it arrives at the STUN server. The STUN server will then 1856 return this IP address in the XOR-MAPPED-ADDRESS attribute to the 1857 client, and send the response back to that (falsified) IP address and 1858 port. If the attacker can also intercept this response, it can 1859 direct it back towards the client. Protecting against this attack by 1860 using a message-integrity check is impossible, since a message- 1861 integrity value cannot cover the source IP address, since the 1862 intervening NAT must be able to modify this value. Instead, one 1863 solution to preventing the attacks listed below is for the client to 1864 verify the reflexive address learned, as is done in ICE 1865 [I-D.ietf-mmusic-ice]. Other usages may use other means to prevent 1866 these attacks. 1868 16.2.1. Attack I: DDoS Against a Target 1870 In this attack, the attacker provides one or more clients with the 1871 same faked reflexive address that points to the intended target. 1872 This will trick the STUN clients into thinking that their reflexive 1873 addresses are equal to that of the target. If the clients hand out 1874 that reflexive address in order to receive traffic on it (for 1875 example, in SIP messages), the traffic will instead be sent to the 1876 target. This attack can provide substantial amplification, 1877 especially when used with clients that are using STUN to enable 1878 multimedia applications. However, it can only be launched against 1879 targets for which packets from the STUN server to the target pass 1880 through the attacker, limiting the cases in which it is possible 1882 16.2.2. Attack II: Silencing a Client 1884 In this attack, the attacker provides a STUN client with a faked 1885 reflexive address. The reflexive address it provides is a transport 1886 address that routes to nowhere. As a result, the client won't 1887 receive any of the packets it expects to receive when it hands out 1888 the reflexive address. This exploitation is not very interesting for 1889 the attacker. It impacts a single client, which is frequently not 1890 the desired target. Moreover, any attacker that can mount the attack 1891 could also deny service to the client by other means, such as 1892 preventing the client from receiving any response from the STUN 1893 server, or even a DHCP server. As with the attack in Section 16.2.1, 1894 this attack is only possible when the attacker is on path for packets 1895 sent from the STUN server towards this unused IP address. 1897 16.2.3. Attack III: Assuming the Identity of a Client 1899 This attack is similar to attack II. However, the faked reflexive 1900 address points to the attacker itself. This allows the attacker to 1901 receive traffic which was destined for the client. 1903 16.2.4. Attack IV: Eavesdropping 1905 In this attack, the attacker forces the client to use a reflexive 1906 address that routes to itself. It then forwards any packets it 1907 receives to the client. This attack would allow the attacker to 1908 observe all packets sent to the client. However, in order to launch 1909 the attack, the attacker must have already been able to observe 1910 packets from the client to the STUN server. In most cases (such as 1911 when the attack is launched from an access network), this means that 1912 the attacker could already observe packets sent to the client. This 1913 attack is, as a result, only useful for observing traffic by 1914 attackers on the path from the client to the STUN server, but not 1915 generally on the path of packets being routed towards the client. 1917 16.3. Hash Agility Plan 1919 This specification uses HMAC-SHA-1 for computation of the message 1920 integrity. If, at a later time, HMAC-SHA-1 is found to be 1921 compromised, the following is the remedy that will be applied. 1923 We will define a STUN extension which introduces a new message 1924 integrity attribute, computed using a new hash. Clients would be 1925 required to include both the new and old message integrity attributes 1926 in their requests or indications. A new server will utilize the new 1927 message integrity attribute, and an old one, the old. After a 1928 transition period where mixed implementations are in deployment, the 1929 old message-integrity attribute will be deprecated by another 1930 specification, and clients will cease including it in requests. 1932 17. IAB Considerations 1934 The IAB has studied the problem of "Unilateral Self Address Fixing" 1935 (UNSAF), which is the general process by which a client attempts to 1936 determine its address in another realm on the other side of a NAT 1937 through a collaborative protocol reflection mechanism (RFC3424 1938 [RFC3424]). STUN can be used to perform this function using a 1939 Binding Request/Response transaction if one agent is behind a NAT and 1940 the other is on the public side of the NAT. 1942 The IAB has mandated that protocols developed for this purpose 1943 document a specific set of considerations. Because some STUN usages 1944 provide UNSAF functions (such as ICE [I-D.ietf-mmusic-ice] ), and 1945 others do not (such as SIP Outbound [I-D.ietf-sip-outbound]), answers 1946 to these considerations need to be addressed by the usages 1947 themselves. 1949 18. IANA Considerations 1951 IANA is hereby requested to create three new registries: a "STUN 1952 Methods Registry", a "STUN Attributes Registry", and a "STUN Error 1953 Codes registry". IANA is also requested to change the name of the 1954 assigned IANA port for STUN from "nat-stun-port" to "stun". 1956 18.1. STUN Methods Registry 1958 A STUN method is a hex number in the range 0x000 - 0x3FF. The 1959 encoding of STUN method into a STUN message is described in 1960 Section 6. 1962 The initial STUN methods are: 1964 0x000: (Reserved) 1965 0x001: Binding 1966 0x002: (Reserved; was SharedSecret) 1968 STUN methods in the range 0x000 - 0x1FF are assigned by IETF Review 1969 [RFC5226]. STUN methods in the range 0x200 - 0x3FF are assigned by 1970 Designated Expert [RFC5226] 1972 18.2. STUN Attribute Registry 1974 A STUN Attribute type is a hex number in the range 0x0000 - 0xFFFF. 1975 STUN attribute types in the range 0x0000 - 0x7FFF are considered 1976 comprehension-required; STUN attribute types in the range 0x8000 - 1977 0xFFFF are considered comprehension-optional. A STUN agent handles 1978 unknown comprehension-required and comprehension-optional attributes 1979 differently. 1981 The initial STUN Attributes types are: 1983 Comprehension-required range (0x0000-0x7FFF): 1984 0x0000: (Reserved) 1985 0x0001: MAPPED-ADDRESS 1986 0x0002: (Reserved; was RESPONSE-ADDRESS) 1987 0x0004: (Reserved; was SOURCE-ADDRESS) 1988 0x0005: (Reserved; was CHANGED-ADDRESS) 1989 0x0006: USERNAME 1990 0x0007: (Reserved; was PASSWORD) 1991 0x0008: MESSAGE-INTEGRITY 1992 0x0009: ERROR-CODE 1993 0x000A: UNKNOWN-ATTRIBUTES 1994 0x000B: (Reserved; was REFLECTED-FROM) 1995 0x0014: REALM 1996 0x0015: NONCE 1997 0x0020: XOR-MAPPED-ADDRESS 1999 Comprehension-optional range (0x8000-0xFFFF) 2000 0x8022: SERVER 2001 0x8023: ALTERNATE-SERVER 2002 0x8028: FINGERPRINT 2003 0x8030: CLIENT 2005 STUN Attribute types in the first half of the comprehension-required 2006 range (0x0000 - 0x3FFF) and in the first half of the comprehension- 2007 optional range (0x8000 - 0xBFFF) are assigned by IETF Review 2009 [RFC5226]. STUN Attribute types in the second half of the 2010 comprehension-required range (0x4000 - 0x7FFF) and in the second half 2011 of the comprehension-optional range (0xC000 - 0xFFFF) are assigned by 2012 Designated Expert [RFC5226]. The responsibility of the expert is to 2013 verify that the selected codepoint(s) are not in use, and that the 2014 request is not for an abnormally large number of codepoints. 2015 Technical review of the extension itself is outside the scope of the 2016 designated expert responsibility. 2018 18.3. STUN Error Code Registry 2020 A STUN Error code is a number in the range 0 - 699. STUN error codes 2021 are accompanied by a textual reason phrase in UTF-8 [RFC3629] which 2022 is intended only for human consumption and can be anything 2023 appropriate; this document proposes only suggested values. 2025 STUN error codes are consistent in codepoint assignments and 2026 semantics with SIP [RFC3261] and HTTP [RFC2616]. 2028 The initial values in this registry are given in Section 15.6. 2030 New STUN error codes are assigned based on IETF Review [RFC5226]. 2031 The specification must carefully consider how clients that do not 2032 understand this error code will process it before granting the 2033 request. See the rules in Section 7.3.4. 2035 18.4. STUN UDP and TCP Port Numbers 2037 IANA has previously assigned port 3478 for STUN. This port appears 2038 in the IANA registry under the moniker "nat-stun-port". In order to 2039 align the DNS SRV procedures with the registered protocol service, 2040 IANA is requested to change the name of protocol assigned to port 2041 3478 from "nat-stun-port" to "stun", and the textual name from 2042 "Simple Traversal of UDP Through NAT (STUN)" to "Session Traversal 2043 Utilities for NAT", so that the IANA port registry would read: 2045 stun 3478/tcp Session Traversal Utilities for NAT (STUN) port 2046 stun 3478/udp Session Traversal Utilities for NAT (STUN) port 2048 In addition, IANA is requested to assign port numbers for the "stuns" 2049 service, defined over TCP and UDP. The UDP port is not currently 2050 defined however is reserved for future use. 2052 19. Changes Since RFC 3489 2054 This specification obsoletes RFC3489 [RFC3489]. This specification 2055 differs from RFC3489 in the following ways: 2057 o Removed the notion that STUN is a complete NAT traversal solution. 2058 STUN is now a tool that can be used to produce a NAT traversal 2059 solution. As a consequence, changed the name of the protocol to 2060 Session Traversal Utilities for NAT. 2062 o Introduced the concept of STUN usages, and described what a usage 2063 of STUN must document. 2065 o Removed the usage of STUN for NAT type detection and binding 2066 lifetime discovery. These techniques have proven overly brittle 2067 due to wider variations in the types of NAT devices than described 2068 in this document. Removed the RESPONSE-ADDRESS, CHANGED-ADDRESS, 2069 CHANGE-REQUEST, SOURCE-ADDRESS, and REFLECTED-FROM attributes. 2071 o Added a fixed 32-bit magic cookie and reduced length of 2072 transaction ID by 32 bits. The magic cookie begins at the same 2073 offset as the original transaction ID. 2075 o Added the XOR-MAPPED-ADDRESS attribute, which is included in 2076 Binding Responses if the magic cookie is present in the request. 2077 Otherwise the RFC3489 behavior is retained (that is, Binding 2078 Response includes MAPPED-ADDRESS). See discussion in XOR-MAPPED- 2079 ADDRESS regarding this change. 2081 o Introduced formal structure into the Message Type header field, 2082 with an explicit pair of bits for indication of request, response, 2083 error response or indication. Consequently, the message type 2084 field is split into the class (one of the previous four) and 2085 method. 2087 o Explicitly point out that the most significant two bits of STUN 2088 are 0b00, allowing easy differentiation with RTP packets when used 2089 with ICE. 2091 o Added the FINGERPRINT attribute to provide a method of definitely 2092 detecting the difference between STUN and another protocol when 2093 the two protocols are multiplexed together. 2095 o Added support for IPv6. Made it clear that an IPv4 client could 2096 get a v6 mapped address, and vice-a-versa. 2098 o Added long-term credential-based authentication. 2100 o Added the SERVER, REALM, NONCE, and ALTERNATE-SERVER attributes. 2102 o Removed the SharedSecret method, and thus the PASSWORD attribute. 2103 This method was almost never implemented and is not needed with 2104 current usages. 2106 o Removed recommendation to continue listening for STUN Responses 2107 for 10 seconds in an attempt to recognize an attack. 2109 o Changed transaction timers to be more TCP friendly. 2111 o Removed the STUN example that centered around the separation of 2112 the control and media planes. Instead, provided more information 2113 on using STUN with protocols. 2115 o Defined a generic padding mechanism that changes the 2116 interpretation of the length attribute. This would, in theory, 2117 break backwards compatibility. However, the mechanism in RFC 3489 2118 never worked for the few attributes that weren't aligned naturally 2119 on 32 bit boundaries. 2121 o REALM, SERVER, reason phrases and NONCE limited to 127 characters. 2122 USERNAME to 513 bytes. 2124 o Changed the DNS SRV procedures for TCP and TLS. UDP remains the 2125 same as before. 2127 20. Contributors 2129 Christian Huitema and Joel Weinberger were original co-authors of RFC 2130 3489. 2132 21. Acknowledgements 2134 The authors would like to thank Cedric Aoun, Pete Cordell, Cullen 2135 Jennings, Bob Penfield, Xavier Marjou, Magnus Westerlund, Miguel 2136 Garcia, Bruce Lowekamp and Chris Sullivan for their comments, and 2137 Baruch Sterman and Alan Hawrylyshen for initial implementations. 2138 Thanks for Leslie Daigle, Allison Mankin, Eric Rescorla, and Henning 2139 Schulzrinne for IESG and IAB input on this work. 2141 22. References 2143 22.1. Normative References 2145 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2146 Requirement Levels", BCP 14, RFC 2119, March 1997. 2148 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 2149 September 1981. 2151 [RFC2782] Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS RR for 2152 specifying the location of services (DNS SRV)", RFC 2782, 2153 February 2000. 2155 [RFC2818] Rescorla, E., "HTTP Over TLS", RFC 2818, May 2000. 2157 [RFC1122] Braden, R., "Requirements for Internet Hosts - 2158 Communication Layers", STD 3, RFC 1122, October 1989. 2160 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2161 (IPv6) Specification", RFC 2460, December 1998. 2163 [RFC2617] Franks, J., Hallam-Baker, P., Hostetler, J., Lawrence, S., 2164 Leach, P., Luotonen, A., and L. Stewart, "HTTP 2165 Authentication: Basic and Digest Access Authentication", 2166 RFC 2617, June 1999. 2168 [RFC2988] Paxson, V. and M. Allman, "Computing TCP's Retransmission 2169 Timer", RFC 2988, November 2000. 2171 [RFC2104] Krawczyk, H., Bellare, M., and R. Canetti, "HMAC: Keyed- 2172 Hashing for Message Authentication", RFC 2104, 2173 February 1997. 2175 [ITU.V42.2002] 2176 International Telecommunications Union, "Error-correcting 2177 Procedures for DCEs Using Asynchronous-to-Synchronous 2178 Conversion", ITU-T Recommendation V.42, March 2002. 2180 [RFC3629] Yergeau, F., "UTF-8, a transformation format of ISO 2181 10646", STD 63, RFC 3629, November 2003. 2183 [RFC1321] Rivest, R., "The MD5 Message-Digest Algorithm", RFC 1321, 2184 April 1992. 2186 [RFC4013] Zeilenga, K., "SASLprep: Stringprep Profile for User Names 2187 and Passwords", RFC 4013, February 2005. 2189 22.2. Informational References 2191 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 2192 A., Peterson, J., Sparks, R., Handley, M., and E. 2193 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 2194 June 2002. 2196 [RFC2616] Fielding, R., Gettys, J., Mogul, J., Frystyk, H., 2197 Masinter, L., Leach, P., and T. Berners-Lee, "Hypertext 2198 Transfer Protocol -- HTTP/1.1", RFC 2616, June 1999. 2200 [RFC4107] Bellovin, S. and R. Housley, "Guidelines for Cryptographic 2201 Key Management", BCP 107, RFC 4107, June 2005. 2203 [I-D.ietf-mmusic-ice] 2204 Rosenberg, J., "Interactive Connectivity Establishment 2205 (ICE): A Protocol for Network Address Translator (NAT) 2206 Traversal for Offer/Answer Protocols", 2207 draft-ietf-mmusic-ice-19 (work in progress), October 2007. 2209 [RFC3489] Rosenberg, J., Weinberger, J., Huitema, C., and R. Mahy, 2210 "STUN - Simple Traversal of User Datagram Protocol (UDP) 2211 Through Network Address Translators (NATs)", RFC 3489, 2212 March 2003. 2214 [I-D.ietf-behave-turn] 2215 Rosenberg, J., Mahy, R., and P. Matthews, "Traversal Using 2216 Relays around NAT (TURN): Relay Extensions to Session 2217 Traversal Utilities for NAT (STUN)", 2218 draft-ietf-behave-turn-06 (work in progress), 2219 January 2008. 2221 [I-D.ietf-sip-outbound] 2222 Jennings, C. and R. Mahy, "Managing Client Initiated 2223 Connections in the Session Initiation Protocol (SIP)", 2224 draft-ietf-sip-outbound-11 (work in progress), 2225 November 2007. 2227 [I-D.ietf-behave-nat-behavior-discovery] 2228 MacDonald, D. and B. Lowekamp, "NAT Behavior Discovery 2229 Using STUN", draft-ietf-behave-nat-behavior-discovery-02 2230 (work in progress), November 2007. 2232 [I-D.ietf-mmusic-ice-tcp] 2233 Rosenberg, J., "TCP Candidates with Interactive 2234 Connectivity Establishment (ICE)", 2235 draft-ietf-mmusic-ice-tcp-05 (work in progress), 2236 November 2007. 2238 [RFC3264] Rosenberg, J. and H. Schulzrinne, "An Offer/Answer Model 2239 with Session Description Protocol (SDP)", RFC 3264, 2240 June 2002. 2242 [RFC3424] Daigle, L. and IAB, "IAB Considerations for UNilateral 2243 Self-Address Fixing (UNSAF) Across Network Address 2244 Translation", RFC 3424, November 2002. 2246 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 2247 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 2248 May 2008. 2250 [KARN87] Karn, P. and C. Partridge, "Improving Round-Trip Time 2251 Estimates in Reliable Transport Protocols", SIGCOMM 1987, 2252 August 1987. 2254 Appendix A. C Snippet to Determine STUN Message Types 2256 Given an 16-bit STUN message type value in host byte order in 2257 msg_type parameter, below are C macros to determine the STUN message 2258 types: 2260 #define IS_REQUEST(msg_type) (((msg_type) & 0x0110) == 0x0000) 2261 #define IS_INDICATION(msg_type) (((msg_type) & 0x0110) == 0x0010) 2262 #define IS_SUCCESS_RESP(msg_type) (((msg_type) & 0x0110) == 0x0100) 2263 #define IS_ERR_RESP(msg_type) (((msg_type) & 0x0110) == 0x0110) 2265 Authors' Addresses 2267 Jonathan Rosenberg 2268 Cisco 2269 Edison, NJ 2270 US 2272 Email: jdrosen@cisco.com 2273 URI: http://www.jdrosen.net 2275 Rohan Mahy 2276 Plantronics 2277 345 Encinal Street 2278 Santa Cruz, CA 95060 2279 US 2281 Email: rohan@ekabal.com 2282 Philip Matthews 2283 Avaya 2284 1135 Innovation Drive 2285 Ottawa, Ontario K2K 3G7 2286 Canada 2288 Phone: +1 613 592 4343 x224 2289 Fax: 2290 Email: philip_matthews@magma.ca 2291 URI: 2293 Dan Wing 2294 Cisco 2295 771 Alder Drive 2296 San Jose, CA 95035 2297 US 2299 Email: dwing@cisco.com 2301 Full Copyright Statement 2303 Copyright (C) The IETF Trust (2008). 2305 This document is subject to the rights, licenses and restrictions 2306 contained in BCP 78, and except as set forth therein, the authors 2307 retain all their rights. 2309 This document and the information contained herein are provided on an 2310 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 2311 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 2312 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 2313 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 2314 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 2315 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 2317 Intellectual Property 2319 The IETF takes no position regarding the validity or scope of any 2320 Intellectual Property Rights or other rights that might be claimed to 2321 pertain to the implementation or use of the technology described in 2322 this document or the extent to which any license under such rights 2323 might or might not be available; nor does it represent that it has 2324 made any independent effort to identify any such rights. Information 2325 on the procedures with respect to rights in RFC documents can be 2326 found in BCP 78 and BCP 79. 2328 Copies of IPR disclosures made to the IETF Secretariat and any 2329 assurances of licenses to be made available, or the result of an 2330 attempt made to obtain a general license or permission for the use of 2331 such proprietary rights by implementers or users of this 2332 specification can be obtained from the IETF on-line IPR repository at 2333 http://www.ietf.org/ipr. 2335 The IETF invites any interested party to bring to its attention any 2336 copyrights, patents or patent applications, or other proprietary 2337 rights that may cover technology that may be required to implement 2338 this standard. Please address the information to the IETF at 2339 ietf-ipr@ietf.org. 2341 Acknowledgment 2343 Funding for the RFC Editor function is provided by the IETF 2344 Administrative Support Activity (IASA).